id	sid	tid	token	lemma	pos
ajst-30200	1	1	academic	academic	ADJ
ajst-30200	1	2	journal	journal	NOUN
ajst-30200	1	3	of	of	ADP
ajst-30200	1	4	science	science	NOUN
ajst-30200	1	5	and	and	CCONJ
ajst-30200	1	6	technology	technology	NOUN
ajst-30200	1	7	issn	issn	NOUN
ajst-30200	1	8	:	:	PUNCT
ajst-30200	1	9	2771	2771	NUM
ajst-30200	1	10	-	-	SYM
ajst-30200	1	11	3032	3032	NUM
ajst-30200	1	12	|	|	NOUN
ajst-30200	1	13	vol	vol	NOUN
ajst-30200	1	14	.	.	PUNCT
ajst-30200	2	1	14	14	NUM
ajst-30200	2	2	,	,	PUNCT
ajst-30200	2	3	no	no	INTJ
ajst-30200	2	4	.	.	NOUN
ajst-30200	2	5	3	3	NUM
ajst-30200	2	6	,	,	PUNCT
ajst-30200	2	7	2025	2025	NUM
ajst-30200	2	8	198	198	NUM
ajst-30200	2	9	adaptive	adaptive	ADJ
ajst-30200	2	10	control	control	NOUN
ajst-30200	2	11	strategy	strategy	NOUN
ajst-30200	2	12	for	for	ADP
ajst-30200	2	13	oscillating	oscillate	VERB
ajst-30200	2	14	water	water	NOUN
ajst-30200	2	15	column	column	NOUN
ajst-30200	2	16	wave	wave	VERB
ajst-30200	2	17	energy	energy	NOUN
ajst-30200	2	18	device	device	NOUN
ajst-30200	2	19	based	base	VERB
ajst-30200	2	20	on	on	ADP
ajst-30200	2	21	deep	deep	ADJ
ajst-30200	2	22	reinforcement	reinforcement	NOUN
ajst-30200	2	23	learning	learn	VERB
ajst-30200	2	24	yuxuan	yuxuan	PROPN
ajst-30200	2	25	liu	liu	PROPN
ajst-30200	2	26	dalian	dalian	PROPN
ajst-30200	2	27	university	university	PROPN
ajst-30200	2	28	of	of	ADP
ajst-30200	2	29	technology	technology	PROPN
ajst-30200	2	30	,	,	PUNCT
ajst-30200	2	31	dalian	dalian	PROPN
ajst-30200	2	32	,	,	PUNCT
ajst-30200	2	33	china	china	PROPN
ajst-30200	2	34	abstract	abstract	NOUN
ajst-30200	2	35	:	:	PUNCT
ajst-30200	2	36	this	this	DET
ajst-30200	2	37	paper	paper	NOUN
ajst-30200	2	38	proposes	propose	VERB
ajst-30200	2	39	an	an	DET
ajst-30200	2	40	adaptive	adaptive	ADJ
ajst-30200	2	41	control	control	NOUN
ajst-30200	2	42	strategy	strategy	NOUN
ajst-30200	2	43	for	for	ADP
ajst-30200	2	44	oscillating	oscillate	VERB
ajst-30200	2	45	water	water	NOUN
ajst-30200	2	46	column	column	NOUN
ajst-30200	2	47	(	(	PUNCT
ajst-30200	2	48	owc	owc	NOUN
ajst-30200	2	49	)	)	PUNCT
ajst-30200	2	50	wave	wave	NOUN
ajst-30200	2	51	energy	energy	NOUN
ajst-30200	2	52	devices	device	NOUN
ajst-30200	2	53	based	base	VERB
ajst-30200	2	54	on	on	ADP
ajst-30200	2	55	deep	deep	ADJ
ajst-30200	2	56	reinforcement	reinforcement	NOUN
ajst-30200	2	57	learning	learning	NOUN
ajst-30200	2	58	(	(	PUNCT
ajst-30200	2	59	drl	drl	PROPN
ajst-30200	2	60	)	)	PUNCT
ajst-30200	2	61	.	.	PUNCT
ajst-30200	3	1	by	by	ADP
ajst-30200	3	2	combining	combine	VERB
ajst-30200	3	3	the	the	DET
ajst-30200	3	4	feature	feature	NOUN
ajst-30200	3	5	extraction	extraction	NOUN
ajst-30200	3	6	capability	capability	NOUN
ajst-30200	3	7	of	of	ADP
ajst-30200	3	8	deep	deep	ADJ
ajst-30200	3	9	neural	neural	ADJ
ajst-30200	3	10	networks	network	NOUN
ajst-30200	3	11	with	with	ADP
ajst-30200	3	12	the	the	DET
ajst-30200	3	13	decision	decision	NOUN
ajst-30200	3	14	optimization	optimization	NOUN
ajst-30200	3	15	ability	ability	NOUN
ajst-30200	3	16	of	of	ADP
ajst-30200	3	17	reinforcement	reinforcement	NOUN
ajst-30200	3	18	learning	learning	NOUN
ajst-30200	3	19	,	,	PUNCT
ajst-30200	3	20	a	a	DET
ajst-30200	3	21	system	system	NOUN
ajst-30200	3	22	is	be	AUX
ajst-30200	3	23	designed	design	VERB
ajst-30200	3	24	to	to	PART
ajst-30200	3	25	dynamically	dynamically	ADV
ajst-30200	3	26	adjust	adjust	VERB
ajst-30200	3	27	the	the	DET
ajst-30200	3	28	turbine	turbine	NOUN
ajst-30200	3	29	control	control	NOUN
ajst-30200	3	30	strategy	strategy	NOUN
ajst-30200	3	31	under	under	ADP
ajst-30200	3	32	varying	vary	VERB
ajst-30200	3	33	wave	wave	NOUN
ajst-30200	3	34	conditions	condition	NOUN
ajst-30200	3	35	.	.	PUNCT
ajst-30200	4	1	experimental	experimental	ADJ
ajst-30200	4	2	results	result	NOUN
ajst-30200	4	3	show	show	VERB
ajst-30200	4	4	that	that	SCONJ
ajst-30200	4	5	the	the	DET
ajst-30200	4	6	proposed	propose	VERB
ajst-30200	4	7	method	method	NOUN
ajst-30200	4	8	effectively	effectively	ADV
ajst-30200	4	9	improves	improve	VERB
ajst-30200	4	10	the	the	DET
ajst-30200	4	11	energy	energy	NOUN
ajst-30200	4	12	conversion	conversion	NOUN
ajst-30200	4	13	efficiency	efficiency	NOUN
ajst-30200	4	14	of	of	ADP
ajst-30200	4	15	the	the	DET
ajst-30200	4	16	device	device	NOUN
ajst-30200	4	17	,	,	PUNCT
ajst-30200	4	18	enhances	enhance	VERB
ajst-30200	4	19	its	its	PRON
ajst-30200	4	20	stability	stability	NOUN
ajst-30200	4	21	under	under	ADP
ajst-30200	4	22	extreme	extreme	ADJ
ajst-30200	4	23	sea	sea	NOUN
ajst-30200	4	24	conditions	condition	NOUN
ajst-30200	4	25	,	,	PUNCT
ajst-30200	4	26	and	and	CCONJ
ajst-30200	4	27	significantly	significantly	ADV
ajst-30200	4	28	optimizes	optimize	VERB
ajst-30200	4	29	the	the	DET
ajst-30200	4	30	system	system	NOUN
ajst-30200	4	31	's	's	PART
ajst-30200	4	32	long	long	ADJ
ajst-30200	4	33	-	-	PUNCT
ajst-30200	4	34	term	term	NOUN
ajst-30200	4	35	operational	operational	ADJ
ajst-30200	4	36	performance	performance	NOUN
ajst-30200	4	37	.	.	PUNCT
ajst-30200	5	1	the	the	DET
ajst-30200	5	2	study	study	NOUN
ajst-30200	5	3	demonstrates	demonstrate	VERB
ajst-30200	5	4	that	that	SCONJ
ajst-30200	5	5	deep	deep	ADJ
ajst-30200	5	6	reinforcement	reinforcement	NOUN
ajst-30200	5	7	learning	learning	NOUN
ajst-30200	5	8	provides	provide	VERB
ajst-30200	5	9	a	a	DET
ajst-30200	5	10	feasible	feasible	ADJ
ajst-30200	5	11	solution	solution	NOUN
ajst-30200	5	12	for	for	ADP
ajst-30200	5	13	intelligent	intelligent	ADJ
ajst-30200	5	14	control	control	NOUN
ajst-30200	5	15	of	of	ADP
ajst-30200	5	16	wave	wave	NOUN
ajst-30200	5	17	energy	energy	NOUN
ajst-30200	5	18	devices	device	NOUN
ajst-30200	5	19	.	.	PUNCT
ajst-30200	6	1	keywords	keyword	NOUN
ajst-30200	6	2	:	:	PUNCT
ajst-30200	6	3	deep	deep	ADJ
ajst-30200	6	4	reinforcement	reinforcement	NOUN
ajst-30200	6	5	learning	learning	NOUN
ajst-30200	6	6	;	;	PUNCT
ajst-30200	6	7	oscillating	oscillate	VERB
ajst-30200	6	8	water	water	NOUN
ajst-30200	6	9	column	column	NOUN
ajst-30200	6	10	;	;	PUNCT
ajst-30200	6	11	wave	wave	NOUN
ajst-30200	6	12	energy	energy	NOUN
ajst-30200	6	13	;	;	PUNCT
ajst-30200	6	14	adaptive	adaptive	ADJ
ajst-30200	6	15	control	control	NOUN
ajst-30200	6	16	;	;	PUNCT
ajst-30200	6	17	energy	energy	NOUN
ajst-30200	6	18	conversion	conversion	NOUN
ajst-30200	6	19	efficiency	efficiency	NOUN
ajst-30200	6	20	.	.	PUNCT
ajst-30200	7	1	1	1	X
ajst-30200	7	2	.	.	X
ajst-30200	7	3	modeling	modeling	NOUN
ajst-30200	7	4	and	and	CCONJ
ajst-30200	7	5	analysis	analysis	NOUN
ajst-30200	7	6	of	of	ADP
ajst-30200	7	7	oscillating	oscillate	VERB
ajst-30200	7	8	water	water	NOUN
ajst-30200	7	9	column	column	NOUN
ajst-30200	7	10	wave	wave	NOUN
ajst-30200	7	11	energy	energy	NOUN
ajst-30200	7	12	device	device	NOUN
ajst-30200	7	13	(	(	PUNCT
ajst-30200	7	14	a	a	X
ajst-30200	7	15	)	)	PUNCT
ajst-30200	7	16	basic	basic	ADJ
ajst-30200	7	17	structure	structure	NOUN
ajst-30200	7	18	and	and	CCONJ
ajst-30200	7	19	working	working	NOUN
ajst-30200	7	20	principle	principle	NOUN
ajst-30200	7	21	of	of	ADP
ajst-30200	7	22	oscillating	oscillate	VERB
ajst-30200	7	23	water	water	NOUN
ajst-30200	7	24	column	column	NOUN
ajst-30200	7	25	wave	wave	VERB
ajst-30200	7	26	energy	energy	NOUN
ajst-30200	7	27	device	device	NOUN
ajst-30200	7	28	1	1	NUM
ajst-30200	7	29	.	.	PUNCT
ajst-30200	7	30	structure	structure	NOUN
ajst-30200	7	31	and	and	CCONJ
ajst-30200	7	32	mechanism	mechanism	NOUN
ajst-30200	7	33	of	of	ADP
ajst-30200	7	34	the	the	DET
ajst-30200	7	35	oscillating	oscillate	VERB
ajst-30200	7	36	water	water	NOUN
ajst-30200	7	37	column	column	NOUN
ajst-30200	7	38	an	an	DET
ajst-30200	7	39	oscillating	oscillate	VERB
ajst-30200	7	40	water	water	NOUN
ajst-30200	7	41	column	column	NOUN
ajst-30200	7	42	(	(	PUNCT
ajst-30200	7	43	owc	owc	NOUN
ajst-30200	7	44	)	)	PUNCT
ajst-30200	7	45	wave	wave	NOUN
ajst-30200	7	46	energy	energy	NOUN
ajst-30200	7	47	device	device	NOUN
ajst-30200	7	48	is	be	AUX
ajst-30200	7	49	a	a	DET
ajst-30200	7	50	system	system	NOUN
ajst-30200	7	51	that	that	PRON
ajst-30200	7	52	utilizes	utilize	VERB
ajst-30200	7	53	the	the	DET
ajst-30200	7	54	up	up	ADP
ajst-30200	7	55	-	-	PUNCT
ajst-30200	7	56	and	and	CCONJ
ajst-30200	7	57	-	-	PUNCT
ajst-30200	7	58	down	down	NOUN
ajst-30200	7	59	motion	motion	NOUN
ajst-30200	7	60	of	of	ADP
ajst-30200	7	61	water	water	NOUN
ajst-30200	7	62	columns	column	NOUN
ajst-30200	7	63	induced	induce	VERB
ajst-30200	7	64	by	by	ADP
ajst-30200	7	65	ocean	ocean	NOUN
ajst-30200	7	66	waves	wave	NOUN
ajst-30200	7	67	to	to	PART
ajst-30200	7	68	drive	drive	VERB
ajst-30200	7	69	a	a	DET
ajst-30200	7	70	gas	gas	NOUN
ajst-30200	7	71	turbine	turbine	NOUN
ajst-30200	7	72	for	for	ADP
ajst-30200	7	73	power	power	NOUN
ajst-30200	7	74	generation	generation	NOUN
ajst-30200	7	75	.	.	PUNCT
ajst-30200	8	1	the	the	DET
ajst-30200	8	2	basic	basic	ADJ
ajst-30200	8	3	structure	structure	NOUN
ajst-30200	8	4	typically	typically	ADV
ajst-30200	8	5	includes	include	VERB
ajst-30200	8	6	a	a	DET
ajst-30200	8	7	closed	closed	ADJ
ajst-30200	8	8	water	water	NOUN
ajst-30200	8	9	tank	tank	NOUN
ajst-30200	8	10	,	,	PUNCT
ajst-30200	8	11	with	with	ADP
ajst-30200	8	12	a	a	DET
ajst-30200	8	13	turbine	turbine	NOUN
ajst-30200	8	14	connected	connect	VERB
ajst-30200	8	15	to	to	ADP
ajst-30200	8	16	the	the	DET
ajst-30200	8	17	outside	outside	ADJ
ajst-30200	8	18	air	air	NOUN
ajst-30200	8	19	via	via	ADP
ajst-30200	8	20	a	a	DET
ajst-30200	8	21	gas	gas	NOUN
ajst-30200	8	22	pipeline	pipeline	NOUN
ajst-30200	8	23	.	.	PUNCT
ajst-30200	9	1	the	the	DET
ajst-30200	9	2	upper	upper	ADJ
ajst-30200	9	3	part	part	NOUN
ajst-30200	9	4	of	of	ADP
ajst-30200	9	5	the	the	DET
ajst-30200	9	6	tank	tank	NOUN
ajst-30200	9	7	is	be	AUX
ajst-30200	9	8	connected	connect	VERB
ajst-30200	9	9	to	to	ADP
ajst-30200	9	10	the	the	DET
ajst-30200	9	11	atmosphere	atmosphere	NOUN
ajst-30200	9	12	,	,	PUNCT
ajst-30200	9	13	while	while	SCONJ
ajst-30200	9	14	the	the	DET
ajst-30200	9	15	lower	low	ADJ
ajst-30200	9	16	part	part	NOUN
ajst-30200	9	17	is	be	AUX
ajst-30200	9	18	in	in	ADP
ajst-30200	9	19	contact	contact	NOUN
ajst-30200	9	20	with	with	ADP
ajst-30200	9	21	seawater	seawater	NOUN
ajst-30200	9	22	.	.	PUNCT
ajst-30200	10	1	when	when	SCONJ
ajst-30200	10	2	waves	wave	NOUN
ajst-30200	10	3	arrive	arrive	VERB
ajst-30200	10	4	,	,	PUNCT
ajst-30200	10	5	the	the	DET
ajst-30200	10	6	seawater	seawater	NOUN
ajst-30200	10	7	causes	cause	VERB
ajst-30200	10	8	vertical	vertical	ADJ
ajst-30200	10	9	oscillations	oscillation	NOUN
ajst-30200	10	10	,	,	PUNCT
ajst-30200	10	11	which	which	PRON
ajst-30200	10	12	drive	drive	VERB
ajst-30200	10	13	the	the	DET
ajst-30200	10	14	water	water	NOUN
ajst-30200	10	15	level	level	NOUN
ajst-30200	10	16	inside	inside	ADP
ajst-30200	10	17	the	the	DET
ajst-30200	10	18	tank	tank	NOUN
ajst-30200	10	19	to	to	PART
ajst-30200	10	20	fluctuate	fluctuate	VERB
ajst-30200	10	21	.	.	PUNCT
ajst-30200	11	1	these	these	DET
ajst-30200	11	2	fluctuations	fluctuation	NOUN
ajst-30200	11	3	compress	compress	VERB
ajst-30200	11	4	and	and	CCONJ
ajst-30200	11	5	expand	expand	VERB
ajst-30200	11	6	the	the	DET
ajst-30200	11	7	air	air	NOUN
ajst-30200	11	8	inside	inside	ADP
ajst-30200	11	9	the	the	DET
ajst-30200	11	10	tank	tank	NOUN
ajst-30200	11	11	,	,	PUNCT
ajst-30200	11	12	thereby	thereby	ADV
ajst-30200	11	13	rotating	rotate	VERB
ajst-30200	11	14	the	the	DET
ajst-30200	11	15	gas	gas	NOUN
ajst-30200	11	16	turbine	turbine	NOUN
ajst-30200	11	17	and	and	CCONJ
ajst-30200	11	18	converting	convert	VERB
ajst-30200	11	19	mechanical	mechanical	ADJ
ajst-30200	11	20	energy	energy	NOUN
ajst-30200	11	21	into	into	ADP
ajst-30200	11	22	electrical	electrical	ADJ
ajst-30200	11	23	energy	energy	NOUN
ajst-30200	11	24	.	.	PUNCT
ajst-30200	12	1	2	2	X
ajst-30200	12	2	.	.	NOUN
ajst-30200	12	3	wave	wave	NOUN
ajst-30200	12	4	energy	energy	NOUN
ajst-30200	12	5	capture	capture	NOUN
ajst-30200	12	6	and	and	CCONJ
ajst-30200	12	7	conversion	conversion	VERB
ajst-30200	12	8	the	the	DET
ajst-30200	12	9	working	work	VERB
ajst-30200	12	10	principle	principle	NOUN
ajst-30200	12	11	of	of	ADP
ajst-30200	12	12	the	the	DET
ajst-30200	12	13	owc	owc	NOUN
ajst-30200	12	14	device	device	NOUN
ajst-30200	12	15	is	be	AUX
ajst-30200	12	16	based	base	VERB
ajst-30200	12	17	on	on	ADP
ajst-30200	12	18	the	the	DET
ajst-30200	12	19	longitudinal	longitudinal	ADJ
ajst-30200	12	20	oscillations	oscillation	NOUN
ajst-30200	12	21	of	of	ADP
ajst-30200	12	22	the	the	DET
ajst-30200	12	23	waves	wave	NOUN
ajst-30200	12	24	.	.	PUNCT
ajst-30200	13	1	these	these	DET
ajst-30200	13	2	oscillations	oscillation	NOUN
ajst-30200	13	3	cause	cause	VERB
ajst-30200	13	4	periodic	periodic	ADJ
ajst-30200	13	5	changes	change	NOUN
ajst-30200	13	6	in	in	ADP
ajst-30200	13	7	the	the	DET
ajst-30200	13	8	water	water	NOUN
ajst-30200	13	9	's	's	PART
ajst-30200	13	10	height	height	NOUN
ajst-30200	13	11	,	,	PUNCT
ajst-30200	13	12	generating	generate	VERB
ajst-30200	13	13	pressure	pressure	NOUN
ajst-30200	13	14	fluctuations	fluctuation	NOUN
ajst-30200	13	15	above	above	ADP
ajst-30200	13	16	the	the	DET
ajst-30200	13	17	water	water	NOUN
ajst-30200	13	18	column	column	NOUN
ajst-30200	13	19	.	.	PUNCT
ajst-30200	14	1	these	these	DET
ajst-30200	14	2	gas	gas	NOUN
ajst-30200	14	3	pressure	pressure	NOUN
ajst-30200	14	4	variations	variation	NOUN
ajst-30200	14	5	drive	drive	VERB
ajst-30200	14	6	the	the	DET
ajst-30200	14	7	turbine	turbine	NOUN
ajst-30200	14	8	,	,	PUNCT
ajst-30200	14	9	which	which	PRON
ajst-30200	14	10	is	be	AUX
ajst-30200	14	11	then	then	ADV
ajst-30200	14	12	converted	convert	VERB
ajst-30200	14	13	into	into	ADP
ajst-30200	14	14	electrical	electrical	ADJ
ajst-30200	14	15	energy	energy	NOUN
ajst-30200	14	16	through	through	ADP
ajst-30200	14	17	a	a	DET
ajst-30200	14	18	generator	generator	NOUN
ajst-30200	14	19	.	.	PUNCT
ajst-30200	15	1	the	the	DET
ajst-30200	15	2	owc	owc	PROPN
ajst-30200	15	3	device	device	NOUN
ajst-30200	15	4	has	have	VERB
ajst-30200	15	5	good	good	ADJ
ajst-30200	15	6	adaptability	adaptability	NOUN
ajst-30200	15	7	in	in	ADP
ajst-30200	15	8	capturing	capture	VERB
ajst-30200	15	9	wave	wave	NOUN
ajst-30200	15	10	energy	energy	NOUN
ajst-30200	15	11	,	,	PUNCT
ajst-30200	15	12	especially	especially	ADV
ajst-30200	15	13	in	in	ADP
ajst-30200	15	14	areas	area	NOUN
ajst-30200	15	15	with	with	ADP
ajst-30200	15	16	fluctuating	fluctuate	VERB
ajst-30200	15	17	sea	sea	NOUN
ajst-30200	15	18	conditions	condition	NOUN
ajst-30200	15	19	.	.	PUNCT
ajst-30200	16	1	the	the	DET
ajst-30200	16	2	core	core	NOUN
ajst-30200	16	3	of	of	ADP
ajst-30200	16	4	the	the	DET
ajst-30200	16	5	device	device	NOUN
ajst-30200	16	6	is	be	AUX
ajst-30200	16	7	to	to	PART
ajst-30200	16	8	effectively	effectively	ADV
ajst-30200	16	9	convert	convert	VERB
ajst-30200	16	10	the	the	DET
ajst-30200	16	11	energy	energy	NOUN
ajst-30200	16	12	of	of	ADP
ajst-30200	16	13	the	the	DET
ajst-30200	16	14	water	water	NOUN
ajst-30200	16	15	column	column	NOUN
ajst-30200	16	16	's	's	PART
ajst-30200	16	17	oscillations	oscillation	NOUN
ajst-30200	16	18	into	into	ADP
ajst-30200	16	19	mechanical	mechanical	ADJ
ajst-30200	16	20	energy	energy	NOUN
ajst-30200	16	21	,	,	PUNCT
ajst-30200	16	22	which	which	PRON
ajst-30200	16	23	is	be	AUX
ajst-30200	16	24	ultimately	ultimately	ADV
ajst-30200	16	25	captured	capture	VERB
ajst-30200	16	26	and	and	CCONJ
ajst-30200	16	27	converted	convert	VERB
ajst-30200	16	28	into	into	ADP
ajst-30200	16	29	usable	usable	ADJ
ajst-30200	16	30	energy[1	energy[1	PROPN
ajst-30200	16	31	]	]	PUNCT
ajst-30200	16	32	.	.	PUNCT
ajst-30200	17	1	3	3	X
ajst-30200	17	2	.	.	X
ajst-30200	17	3	energy	energy	NOUN
ajst-30200	17	4	conversion	conversion	NOUN
ajst-30200	17	5	efficiency	efficiency	NOUN
ajst-30200	17	6	and	and	CCONJ
ajst-30200	17	7	influencing	influence	VERB
ajst-30200	17	8	factors	factor	NOUN
ajst-30200	17	9	energy	energy	NOUN
ajst-30200	17	10	conversion	conversion	NOUN
ajst-30200	17	11	efficiency	efficiency	NOUN
ajst-30200	17	12	is	be	AUX
ajst-30200	17	13	an	an	DET
ajst-30200	17	14	important	important	ADJ
ajst-30200	17	15	indicator	indicator	NOUN
ajst-30200	17	16	of	of	ADP
ajst-30200	17	17	the	the	DET
ajst-30200	17	18	performance	performance	NOUN
ajst-30200	17	19	of	of	ADP
ajst-30200	17	20	the	the	DET
ajst-30200	17	21	owc	owc	PROPN
ajst-30200	17	22	device	device	NOUN
ajst-30200	17	23	.	.	PUNCT
ajst-30200	18	1	factors	factor	NOUN
ajst-30200	18	2	influencing	influence	VERB
ajst-30200	18	3	energy	energy	NOUN
ajst-30200	18	4	conversion	conversion	NOUN
ajst-30200	18	5	efficiency	efficiency	NOUN
ajst-30200	18	6	include	include	VERB
ajst-30200	18	7	wave	wave	NOUN
ajst-30200	18	8	height	height	NOUN
ajst-30200	18	9	,	,	PUNCT
ajst-30200	18	10	frequency	frequency	NOUN
ajst-30200	18	11	,	,	PUNCT
ajst-30200	18	12	wave	wave	NOUN
ajst-30200	18	13	directionality	directionality	NOUN
ajst-30200	18	14	,	,	PUNCT
ajst-30200	18	15	and	and	CCONJ
ajst-30200	18	16	the	the	DET
ajst-30200	18	17	design	design	NOUN
ajst-30200	18	18	parameters	parameter	NOUN
ajst-30200	18	19	of	of	ADP
ajst-30200	18	20	the	the	DET
ajst-30200	18	21	device	device	NOUN
ajst-30200	18	22	itself	itself	PRON
ajst-30200	18	23	.	.	PUNCT
ajst-30200	19	1	for	for	ADP
ajst-30200	19	2	example	example	NOUN
ajst-30200	19	3	,	,	PUNCT
ajst-30200	19	4	when	when	SCONJ
ajst-30200	19	5	the	the	DET
ajst-30200	19	6	wave	wave	NOUN
ajst-30200	19	7	period	period	NOUN
ajst-30200	19	8	matches	match	VERB
ajst-30200	19	9	the	the	DET
ajst-30200	19	10	resonance	resonance	NOUN
ajst-30200	19	11	frequency	frequency	NOUN
ajst-30200	19	12	of	of	ADP
ajst-30200	19	13	the	the	DET
ajst-30200	19	14	owc	owc	PROPN
ajst-30200	19	15	device	device	NOUN
ajst-30200	19	16	,	,	PUNCT
ajst-30200	19	17	a	a	DET
ajst-30200	19	18	higher	high	ADJ
ajst-30200	19	19	energy	energy	NOUN
ajst-30200	19	20	conversion	conversion	NOUN
ajst-30200	19	21	efficiency	efficiency	NOUN
ajst-30200	19	22	can	can	AUX
ajst-30200	19	23	be	be	AUX
ajst-30200	19	24	achieved	achieve	VERB
ajst-30200	19	25	,	,	PUNCT
ajst-30200	19	26	while	while	SCONJ
ajst-30200	19	27	frequency	frequency	NOUN
ajst-30200	19	28	mismatches	mismatch	NOUN
ajst-30200	19	29	lead	lead	VERB
ajst-30200	19	30	to	to	ADP
ajst-30200	19	31	energy	energy	NOUN
ajst-30200	19	32	loss	loss	NOUN
ajst-30200	19	33	.	.	PUNCT
ajst-30200	20	1	the	the	DET
ajst-30200	20	2	size	size	NOUN
ajst-30200	20	3	of	of	ADP
ajst-30200	20	4	the	the	DET
ajst-30200	20	5	water	water	NOUN
ajst-30200	20	6	tank	tank	NOUN
ajst-30200	20	7	,	,	PUNCT
ajst-30200	20	8	the	the	DET
ajst-30200	20	9	design	design	NOUN
ajst-30200	20	10	of	of	ADP
ajst-30200	20	11	the	the	DET
ajst-30200	20	12	turbine	turbine	NOUN
ajst-30200	20	13	,	,	PUNCT
ajst-30200	20	14	and	and	CCONJ
ajst-30200	20	15	the	the	DET
ajst-30200	20	16	configuration	configuration	NOUN
ajst-30200	20	17	of	of	ADP
ajst-30200	20	18	the	the	DET
ajst-30200	20	19	air	air	NOUN
ajst-30200	20	20	intake	intake	NOUN
ajst-30200	20	21	and	and	CCONJ
ajst-30200	20	22	exhaust	exhaust	NOUN
ajst-30200	20	23	ports	port	NOUN
ajst-30200	20	24	are	be	AUX
ajst-30200	20	25	also	also	ADV
ajst-30200	20	26	key	key	ADJ
ajst-30200	20	27	factors	factor	NOUN
ajst-30200	20	28	affecting	affect	VERB
ajst-30200	20	29	energy	energy	NOUN
ajst-30200	20	30	conversion	conversion	NOUN
ajst-30200	20	31	efficiency	efficiency	NOUN
ajst-30200	20	32	.	.	PUNCT
ajst-30200	21	1	additionally	additionally	ADV
ajst-30200	21	2	,	,	PUNCT
ajst-30200	21	3	the	the	DET
ajst-30200	21	4	device	device	NOUN
ajst-30200	21	5	's	's	PART
ajst-30200	21	6	adaptability	adaptability	NOUN
ajst-30200	21	7	to	to	ADP
ajst-30200	21	8	various	various	ADJ
ajst-30200	21	9	sea	sea	NOUN
ajst-30200	21	10	conditions	condition	NOUN
ajst-30200	21	11	directly	directly	ADV
ajst-30200	21	12	influences	influence	VERB
ajst-30200	21	13	its	its	PRON
ajst-30200	21	14	long	long	ADJ
ajst-30200	21	15	-	-	PUNCT
ajst-30200	21	16	term	term	NOUN
ajst-30200	21	17	energy	energy	NOUN
ajst-30200	21	18	output	output	NOUN
ajst-30200	21	19	.	.	PUNCT
ajst-30200	22	1	(	(	PUNCT
ajst-30200	22	2	b	b	X
ajst-30200	22	3	)	)	PUNCT
ajst-30200	22	4	dynamic	dynamic	ADJ
ajst-30200	22	5	model	model	NOUN
ajst-30200	22	6	of	of	ADP
ajst-30200	22	7	the	the	DET
ajst-30200	22	8	oscillating	oscillate	VERB
ajst-30200	22	9	water	water	NOUN
ajst-30200	22	10	column	column	NOUN
ajst-30200	22	11	system	system	NOUN
ajst-30200	22	12	1	1	NUM
ajst-30200	22	13	.	.	PUNCT
ajst-30200	23	1	system	system	NOUN
ajst-30200	23	2	dynamic	dynamic	ADJ
ajst-30200	23	3	modeling	modeling	NOUN
ajst-30200	23	4	method	method	NOUN
ajst-30200	23	5	the	the	DET
ajst-30200	23	6	dynamic	dynamic	ADJ
ajst-30200	23	7	modeling	modeling	NOUN
ajst-30200	23	8	of	of	ADP
ajst-30200	23	9	the	the	DET
ajst-30200	23	10	owc	owc	PROPN
ajst-30200	23	11	wave	wave	NOUN
ajst-30200	23	12	energy	energy	NOUN
ajst-30200	23	13	device	device	NOUN
ajst-30200	23	14	is	be	AUX
ajst-30200	23	15	usually	usually	ADV
ajst-30200	23	16	achieved	achieve	VERB
ajst-30200	23	17	by	by	ADP
ajst-30200	23	18	combining	combine	VERB
ajst-30200	23	19	the	the	DET
ajst-30200	23	20	fluid	fluid	ADJ
ajst-30200	23	21	dynamics	dynamic	NOUN
ajst-30200	23	22	equations	equation	NOUN
ajst-30200	23	23	governing	govern	VERB
ajst-30200	23	24	the	the	DET
ajst-30200	23	25	movement	movement	NOUN
ajst-30200	23	26	of	of	ADP
ajst-30200	23	27	water	water	NOUN
ajst-30200	23	28	with	with	ADP
ajst-30200	23	29	the	the	DET
ajst-30200	23	30	gas	gas	NOUN
ajst-30200	23	31	dynamics	dynamic	NOUN
ajst-30200	23	32	equations	equation	NOUN
ajst-30200	23	33	describing	describe	VERB
ajst-30200	23	34	pressure	pressure	NOUN
ajst-30200	23	35	changes	change	NOUN
ajst-30200	23	36	and	and	CCONJ
ajst-30200	23	37	turbine	turbine	NOUN
ajst-30200	23	38	rotation	rotation	NOUN
ajst-30200	23	39	.	.	PUNCT
ajst-30200	24	1	the	the	DET
ajst-30200	24	2	water	water	NOUN
ajst-30200	24	3	movement	movement	NOUN
ajst-30200	24	4	equations	equation	NOUN
ajst-30200	24	5	are	be	AUX
ajst-30200	24	6	based	base	VERB
ajst-30200	24	7	on	on	ADP
ajst-30200	24	8	the	the	DET
ajst-30200	24	9	fundamental	fundamental	ADJ
ajst-30200	24	10	principles	principle	NOUN
ajst-30200	24	11	of	of	ADP
ajst-30200	24	12	fluid	fluid	ADJ
ajst-30200	24	13	mechanics	mechanic	NOUN
ajst-30200	24	14	,	,	PUNCT
ajst-30200	24	15	considering	consider	VERB
ajst-30200	24	16	the	the	DET
ajst-30200	24	17	propagation	propagation	NOUN
ajst-30200	24	18	and	and	CCONJ
ajst-30200	24	19	reflection	reflection	NOUN
ajst-30200	24	20	of	of	ADP
ajst-30200	24	21	waves	wave	NOUN
ajst-30200	24	22	within	within	ADP
ajst-30200	24	23	the	the	DET
ajst-30200	24	24	device	device	NOUN
ajst-30200	24	25	.	.	PUNCT
ajst-30200	25	1	the	the	DET
ajst-30200	25	2	gas	gas	NOUN
ajst-30200	25	3	dynamics	dynamic	NOUN
ajst-30200	25	4	equations	equation	NOUN
ajst-30200	25	5	describe	describe	VERB
ajst-30200	25	6	the	the	DET
ajst-30200	25	7	pressure	pressure	NOUN
ajst-30200	25	8	variations	variation	NOUN
ajst-30200	25	9	in	in	ADP
ajst-30200	25	10	the	the	DET
ajst-30200	25	11	enclosed	enclose	VERB
ajst-30200	25	12	container	container	NOUN
ajst-30200	25	13	and	and	CCONJ
ajst-30200	25	14	the	the	DET
ajst-30200	25	15	rotational	rotational	ADJ
ajst-30200	25	16	motion	motion	NOUN
ajst-30200	25	17	of	of	ADP
ajst-30200	25	18	the	the	DET
ajst-30200	25	19	turbine	turbine	NOUN
ajst-30200	25	20	.	.	PUNCT
ajst-30200	26	1	a	a	DET
ajst-30200	26	2	common	common	ADJ
ajst-30200	26	3	approach	approach	NOUN
ajst-30200	26	4	is	be	AUX
ajst-30200	26	5	to	to	PART
ajst-30200	26	6	combine	combine	VERB
ajst-30200	26	7	fluid	fluid	ADJ
ajst-30200	26	8	dynamics	dynamic	NOUN
ajst-30200	26	9	and	and	CCONJ
ajst-30200	26	10	aerodynamics	aerodynamic	NOUN
ajst-30200	26	11	models	model	NOUN
ajst-30200	26	12	for	for	ADP
ajst-30200	26	13	accurate	accurate	ADJ
ajst-30200	26	14	simulation	simulation	NOUN
ajst-30200	26	15	.	.	PUNCT
ajst-30200	27	1	2	2	X
ajst-30200	27	2	.	.	X
ajst-30200	27	3	analysis	analysis	NOUN
ajst-30200	27	4	of	of	ADP
ajst-30200	27	5	the	the	DET
ajst-30200	27	6	system	system	NOUN
ajst-30200	27	7	's	's	PART
ajst-30200	27	8	nonlinear	nonlinear	ADJ
ajst-30200	27	9	characteristics	characteristic	NOUN
ajst-30200	27	10	the	the	DET
ajst-30200	27	11	owc	owc	PROPN
ajst-30200	27	12	system	system	NOUN
ajst-30200	27	13	typically	typically	ADV
ajst-30200	27	14	exhibits	exhibit	VERB
ajst-30200	27	15	nonlinear	nonlinear	ADJ
ajst-30200	27	16	characteristics	characteristic	NOUN
ajst-30200	27	17	,	,	PUNCT
ajst-30200	27	18	especially	especially	ADV
ajst-30200	27	19	when	when	SCONJ
ajst-30200	27	20	wave	wave	NOUN
ajst-30200	27	21	amplitudes	amplitude	NOUN
ajst-30200	27	22	are	be	AUX
ajst-30200	27	23	large	large	ADJ
ajst-30200	27	24	.	.	PUNCT
ajst-30200	28	1	the	the	DET
ajst-30200	28	2	oscillations	oscillation	NOUN
ajst-30200	28	3	of	of	ADP
ajst-30200	28	4	the	the	DET
ajst-30200	28	5	water	water	NOUN
ajst-30200	28	6	and	and	CCONJ
ajst-30200	28	7	the	the	DET
ajst-30200	28	8	airflow	airflow	NOUN
ajst-30200	28	9	are	be	AUX
ajst-30200	28	10	interrelated	interrelated	ADJ
ajst-30200	28	11	,	,	PUNCT
ajst-30200	28	12	and	and	CCONJ
ajst-30200	28	13	nonlinear	nonlinear	ADJ
ajst-30200	28	14	dynamics	dynamic	NOUN
ajst-30200	28	15	make	make	VERB
ajst-30200	28	16	the	the	DET
ajst-30200	28	17	system	system	NOUN
ajst-30200	28	18	’s	’s	PART
ajst-30200	28	19	response	response	NOUN
ajst-30200	28	20	depend	depend	VERB
ajst-30200	28	21	not	not	PART
ajst-30200	28	22	only	only	ADV
ajst-30200	28	23	on	on	ADP
ajst-30200	28	24	the	the	DET
ajst-30200	28	25	input	input	NOUN
ajst-30200	28	26	wave	wave	NOUN
ajst-30200	28	27	but	but	CCONJ
ajst-30200	28	28	also	also	ADV
ajst-30200	28	29	on	on	ADP
ajst-30200	28	30	the	the	DET
ajst-30200	28	31	internal	internal	ADJ
ajst-30200	28	32	interactions	interaction	NOUN
ajst-30200	28	33	within	within	ADP
ajst-30200	28	34	the	the	DET
ajst-30200	28	35	system	system	NOUN
ajst-30200	28	36	.	.	PUNCT
ajst-30200	29	1	due	due	ADP
ajst-30200	29	2	to	to	ADP
ajst-30200	29	3	the	the	DET
ajst-30200	29	4	variations	variation	NOUN
ajst-30200	29	5	in	in	ADP
ajst-30200	29	6	wave	wave	NOUN
ajst-30200	29	7	frequency	frequency	NOUN
ajst-30200	29	8	,	,	PUNCT
ajst-30200	29	9	amplitude	amplitude	NOUN
ajst-30200	29	10	,	,	PUNCT
ajst-30200	29	11	and	and	CCONJ
ajst-30200	29	12	the	the	DET
ajst-30200	29	13	dynamic	dynamic	ADJ
ajst-30200	29	14	properties	property	NOUN
ajst-30200	29	15	of	of	ADP
ajst-30200	29	16	the	the	DET
ajst-30200	29	17	water	water	NOUN
ajst-30200	29	18	column	column	NOUN
ajst-30200	29	19	system	system	NOUN
ajst-30200	29	20	,	,	PUNCT
ajst-30200	29	21	a	a	DET
ajst-30200	29	22	nonlinear	nonlinear	ADJ
ajst-30200	29	23	model	model	NOUN
ajst-30200	29	24	is	be	AUX
ajst-30200	29	25	essential	essential	ADJ
ajst-30200	29	26	for	for	ADP
ajst-30200	29	27	accurately	accurately	ADV
ajst-30200	29	28	simulating	simulate	VERB
ajst-30200	29	29	the	the	DET
ajst-30200	29	30	device	device	NOUN
ajst-30200	29	31	's	's	PART
ajst-30200	29	32	actual	actual	ADJ
ajst-30200	29	33	operation	operation	NOUN
ajst-30200	29	34	.	.	PUNCT
ajst-30200	30	1	analyzing	analyze	VERB
ajst-30200	30	2	these	these	DET
ajst-30200	30	3	nonlinear	nonlinear	ADJ
ajst-30200	30	4	characteristics	characteristic	NOUN
ajst-30200	30	5	helps	help	VERB
ajst-30200	30	6	in	in	ADP
ajst-30200	30	7	understanding	understand	VERB
ajst-30200	30	8	the	the	DET
ajst-30200	30	9	performance	performance	NOUN
ajst-30200	30	10	fluctuations	fluctuation	NOUN
ajst-30200	30	11	and	and	CCONJ
ajst-30200	30	12	control	control	NOUN
ajst-30200	30	13	challenges	challenge	NOUN
ajst-30200	30	14	encountered	encounter	VERB
ajst-30200	30	15	during	during	ADP
ajst-30200	30	16	practical	practical	ADJ
ajst-30200	30	17	operation	operation	NOUN
ajst-30200	30	18	.	.	PUNCT
ajst-30200	31	1	3	3	X
ajst-30200	31	2	.	.	X
ajst-30200	31	3	stability	stability	NOUN
ajst-30200	31	4	and	and	CCONJ
ajst-30200	31	5	control	control	NOUN
ajst-30200	31	6	requirements	requirement	NOUN
ajst-30200	31	7	of	of	ADP
ajst-30200	31	8	the	the	DET
ajst-30200	31	9	model	model	NOUN
ajst-30200	31	10	the	the	DET
ajst-30200	31	11	control	control	NOUN
ajst-30200	31	12	requirements	requirement	NOUN
ajst-30200	31	13	for	for	ADP
ajst-30200	31	14	the	the	DET
ajst-30200	31	15	owc	owc	PROPN
ajst-30200	31	16	wave	wave	NOUN
ajst-30200	31	17	energy	energy	NOUN
ajst-30200	31	18	device	device	NOUN
ajst-30200	31	19	mainly	mainly	ADV
ajst-30200	31	20	focus	focus	VERB
ajst-30200	31	21	on	on	ADP
ajst-30200	31	22	how	how	SCONJ
ajst-30200	31	23	to	to	PART
ajst-30200	31	24	effectively	effectively	ADV
ajst-30200	31	25	control	control	VERB
ajst-30200	31	26	the	the	DET
ajst-30200	31	27	turbine	turbine	NOUN
ajst-30200	31	28	and	and	CCONJ
ajst-30200	31	29	handle	handle	VERB
ajst-30200	31	30	the	the	DET
ajst-30200	31	31	instability	instability	NOUN
ajst-30200	31	32	factors	factor	NOUN
ajst-30200	31	33	in	in	ADP
ajst-30200	31	34	the	the	DET
ajst-30200	31	35	wave	wave	NOUN
ajst-30200	31	36	energy	energy	NOUN
ajst-30200	31	37	capture	capture	NOUN
ajst-30200	31	38	process	process	NOUN
ajst-30200	31	39	.	.	PUNCT
ajst-30200	32	1	based	base	VERB
ajst-30200	32	2	on	on	ADP
ajst-30200	32	3	the	the	DET
ajst-30200	32	4	dynamic	dynamic	ADJ
ajst-30200	32	5	model	model	NOUN
ajst-30200	32	6	,	,	PUNCT
ajst-30200	32	7	stability	stability	NOUN
ajst-30200	32	8	analysis	analysis	NOUN
ajst-30200	32	9	helps	help	VERB
ajst-30200	32	10	199	199	NUM
ajst-30200	32	11	evaluate	evaluate	VERB
ajst-30200	32	12	the	the	DET
ajst-30200	32	13	system	system	NOUN
ajst-30200	32	14	's	's	PART
ajst-30200	32	15	performance	performance	NOUN
ajst-30200	32	16	under	under	ADP
ajst-30200	32	17	different	different	ADJ
ajst-30200	32	18	operating	operating	NOUN
ajst-30200	32	19	conditions	condition	NOUN
ajst-30200	32	20	,	,	PUNCT
ajst-30200	32	21	particularly	particularly	ADV
ajst-30200	32	22	considering	consider	VERB
ajst-30200	32	23	the	the	DET
ajst-30200	32	24	potential	potential	ADJ
ajst-30200	32	25	instability	instability	NOUN
ajst-30200	32	26	caused	cause	VERB
ajst-30200	32	27	by	by	ADP
ajst-30200	32	28	nonlinear	nonlinear	ADJ
ajst-30200	32	29	effects	effect	NOUN
ajst-30200	32	30	.	.	PUNCT
ajst-30200	33	1	to	to	PART
ajst-30200	33	2	achieve	achieve	VERB
ajst-30200	33	3	efficient	efficient	ADJ
ajst-30200	33	4	energy	energy	NOUN
ajst-30200	33	5	conversion	conversion	NOUN
ajst-30200	33	6	,	,	PUNCT
ajst-30200	33	7	a	a	DET
ajst-30200	33	8	suitable	suitable	ADJ
ajst-30200	33	9	control	control	NOUN
ajst-30200	33	10	system	system	NOUN
ajst-30200	33	11	must	must	AUX
ajst-30200	33	12	be	be	AUX
ajst-30200	33	13	designed	design	VERB
ajst-30200	33	14	to	to	PART
ajst-30200	33	15	ensure	ensure	VERB
ajst-30200	33	16	stable	stable	ADJ
ajst-30200	33	17	operation	operation	NOUN
ajst-30200	33	18	of	of	ADP
ajst-30200	33	19	the	the	DET
ajst-30200	33	20	device	device	NOUN
ajst-30200	33	21	under	under	ADP
ajst-30200	33	22	varying	vary	VERB
ajst-30200	33	23	wave	wave	NOUN
ajst-30200	33	24	conditions	condition	NOUN
ajst-30200	33	25	.	.	PUNCT
ajst-30200	34	1	this	this	PRON
ajst-30200	34	2	is	be	AUX
ajst-30200	34	3	crucial	crucial	ADJ
ajst-30200	34	4	to	to	PART
ajst-30200	34	5	prevent	prevent	VERB
ajst-30200	34	6	structural	structural	ADJ
ajst-30200	34	7	damage	damage	NOUN
ajst-30200	34	8	or	or	CCONJ
ajst-30200	34	9	performance	performance	NOUN
ajst-30200	34	10	degradation	degradation	NOUN
ajst-30200	34	11	,	,	PUNCT
ajst-30200	34	12	especially	especially	ADV
ajst-30200	34	13	in	in	ADP
ajst-30200	34	14	extreme	extreme	ADJ
ajst-30200	34	15	sea	sea	NOUN
ajst-30200	34	16	conditions	condition	NOUN
ajst-30200	34	17	.	.	PUNCT
ajst-30200	35	1	(	(	PUNCT
ajst-30200	35	2	c	c	X
ajst-30200	35	3	)	)	PUNCT
ajst-30200	35	4	energy	energy	NOUN
ajst-30200	35	5	extraction	extraction	NOUN
ajst-30200	35	6	mechanism	mechanism	NOUN
ajst-30200	35	7	and	and	CCONJ
ajst-30200	35	8	optimization	optimization	NOUN
ajst-30200	35	9	model	model	NOUN
ajst-30200	35	10	1	1	NUM
ajst-30200	35	11	.	.	PUNCT
ajst-30200	36	1	establishing	establish	VERB
ajst-30200	36	2	the	the	DET
ajst-30200	36	3	wave	wave	NOUN
ajst-30200	36	4	energy	energy	NOUN
ajst-30200	36	5	extraction	extraction	NOUN
ajst-30200	36	6	model	model	NOUN
ajst-30200	36	7	the	the	DET
ajst-30200	36	8	wave	wave	NOUN
ajst-30200	36	9	energy	energy	NOUN
ajst-30200	36	10	extraction	extraction	NOUN
ajst-30200	36	11	model	model	NOUN
ajst-30200	36	12	is	be	AUX
ajst-30200	36	13	designed	design	VERB
ajst-30200	36	14	based	base	VERB
ajst-30200	36	15	on	on	ADP
ajst-30200	36	16	the	the	DET
ajst-30200	36	17	dynamic	dynamic	ADJ
ajst-30200	36	18	characteristics	characteristic	NOUN
ajst-30200	36	19	of	of	ADP
ajst-30200	36	20	the	the	DET
ajst-30200	36	21	device	device	NOUN
ajst-30200	36	22	and	and	CCONJ
ajst-30200	36	23	the	the	DET
ajst-30200	36	24	efficiency	efficiency	NOUN
ajst-30200	36	25	of	of	ADP
ajst-30200	36	26	the	the	DET
ajst-30200	36	27	gas	gas	NOUN
ajst-30200	36	28	turbine	turbine	NOUN
ajst-30200	36	29	.	.	PUNCT
ajst-30200	37	1	in	in	ADP
ajst-30200	37	2	the	the	DET
ajst-30200	37	3	model	model	NOUN
ajst-30200	37	4	,	,	PUNCT
ajst-30200	37	5	factors	factor	NOUN
ajst-30200	37	6	such	such	ADJ
ajst-30200	37	7	as	as	ADP
ajst-30200	37	8	wave	wave	NOUN
ajst-30200	37	9	period	period	NOUN
ajst-30200	37	10	,	,	PUNCT
ajst-30200	37	11	wave	wave	NOUN
ajst-30200	37	12	amplitude	amplitude	NOUN
ajst-30200	37	13	,	,	PUNCT
ajst-30200	37	14	and	and	CCONJ
ajst-30200	37	15	direction	direction	NOUN
ajst-30200	37	16	are	be	AUX
ajst-30200	37	17	usually	usually	ADV
ajst-30200	37	18	considered	consider	VERB
ajst-30200	37	19	.	.	PUNCT
ajst-30200	38	1	a	a	DET
ajst-30200	38	2	gas	gas	NOUN
ajst-30200	38	3	pressure	pressure	NOUN
ajst-30200	38	4	fluctuation	fluctuation	NOUN
ajst-30200	38	5	model	model	NOUN
ajst-30200	38	6	within	within	ADP
ajst-30200	38	7	the	the	DET
ajst-30200	38	8	water	water	NOUN
ajst-30200	38	9	column	column	NOUN
ajst-30200	38	10	is	be	AUX
ajst-30200	38	11	established	establish	VERB
ajst-30200	38	12	to	to	PART
ajst-30200	38	13	calculate	calculate	VERB
ajst-30200	38	14	the	the	DET
ajst-30200	38	15	power	power	NOUN
ajst-30200	38	16	that	that	PRON
ajst-30200	38	17	the	the	DET
ajst-30200	38	18	turbine	turbine	NOUN
ajst-30200	38	19	can	can	AUX
ajst-30200	38	20	extract	extract	VERB
ajst-30200	38	21	.	.	PUNCT
ajst-30200	39	1	the	the	DET
ajst-30200	39	2	model	model	NOUN
ajst-30200	39	3	needs	need	VERB
ajst-30200	39	4	to	to	PART
ajst-30200	39	5	account	account	VERB
ajst-30200	39	6	for	for	ADP
ajst-30200	39	7	the	the	DET
ajst-30200	39	8	compression	compression	NOUN
ajst-30200	39	9	and	and	CCONJ
ajst-30200	39	10	expansion	expansion	NOUN
ajst-30200	39	11	effects	effect	NOUN
ajst-30200	39	12	of	of	ADP
ajst-30200	39	13	air	air	NOUN
ajst-30200	39	14	inside	inside	ADP
ajst-30200	39	15	the	the	DET
ajst-30200	39	16	water	water	NOUN
ajst-30200	39	17	tank	tank	NOUN
ajst-30200	39	18	under	under	ADP
ajst-30200	39	19	different	different	ADJ
ajst-30200	39	20	wave	wave	NOUN
ajst-30200	39	21	conditions	condition	NOUN
ajst-30200	39	22	,	,	PUNCT
ajst-30200	39	23	the	the	DET
ajst-30200	39	24	response	response	NOUN
ajst-30200	39	25	of	of	ADP
ajst-30200	39	26	the	the	DET
ajst-30200	39	27	turbine	turbine	NOUN
ajst-30200	39	28	,	,	PUNCT
ajst-30200	39	29	and	and	CCONJ
ajst-30200	39	30	energy	energy	NOUN
ajst-30200	39	31	losses	loss	NOUN
ajst-30200	39	32	.	.	PUNCT
ajst-30200	40	1	by	by	ADP
ajst-30200	40	2	detailing	detail	VERB
ajst-30200	40	3	the	the	DET
ajst-30200	40	4	wave	wave	NOUN
ajst-30200	40	5	energy	energy	NOUN
ajst-30200	40	6	extraction	extraction	NOUN
ajst-30200	40	7	model	model	NOUN
ajst-30200	40	8	,	,	PUNCT
ajst-30200	40	9	the	the	DET
ajst-30200	40	10	performance	performance	NOUN
ajst-30200	40	11	of	of	ADP
ajst-30200	40	12	the	the	DET
ajst-30200	40	13	device	device	NOUN
ajst-30200	40	14	can	can	AUX
ajst-30200	40	15	be	be	AUX
ajst-30200	40	16	accurately	accurately	ADV
ajst-30200	40	17	predicted	predict	VERB
ajst-30200	40	18	,	,	PUNCT
ajst-30200	40	19	providing	provide	VERB
ajst-30200	40	20	a	a	DET
ajst-30200	40	21	theoretical	theoretical	ADJ
ajst-30200	40	22	basis	basis	NOUN
ajst-30200	40	23	for	for	ADP
ajst-30200	40	24	designing	design	VERB
ajst-30200	40	25	the	the	DET
ajst-30200	40	26	control	control	NOUN
ajst-30200	40	27	strategy[2	strategy[2	PROPN
ajst-30200	40	28	]	]	PUNCT
ajst-30200	40	29	.	.	PUNCT
ajst-30200	41	1	2	2	X
ajst-30200	41	2	.	.	X
ajst-30200	41	3	parameter	parameter	NOUN
ajst-30200	41	4	optimization	optimization	NOUN
ajst-30200	41	5	and	and	CCONJ
ajst-30200	41	6	control	control	VERB
ajst-30200	41	7	objective	objective	ADJ
ajst-30200	41	8	setting	setting	NOUN
ajst-30200	41	9	in	in	ADP
ajst-30200	41	10	practical	practical	ADJ
ajst-30200	41	11	applications	application	NOUN
ajst-30200	41	12	,	,	PUNCT
ajst-30200	41	13	the	the	DET
ajst-30200	41	14	performance	performance	NOUN
ajst-30200	41	15	of	of	ADP
ajst-30200	41	16	the	the	DET
ajst-30200	41	17	oscillating	oscillate	VERB
ajst-30200	41	18	water	water	NOUN
ajst-30200	41	19	column	column	NOUN
ajst-30200	41	20	wave	wave	NOUN
ajst-30200	41	21	energy	energy	NOUN
ajst-30200	41	22	device	device	NOUN
ajst-30200	41	23	depends	depend	VERB
ajst-30200	41	24	not	not	PART
ajst-30200	41	25	only	only	ADV
ajst-30200	41	26	on	on	ADP
ajst-30200	41	27	the	the	DET
ajst-30200	41	28	characteristics	characteristic	NOUN
ajst-30200	41	29	of	of	ADP
ajst-30200	41	30	the	the	DET
ajst-30200	41	31	waves	wave	NOUN
ajst-30200	41	32	but	but	CCONJ
ajst-30200	41	33	also	also	ADV
ajst-30200	41	34	on	on	ADP
ajst-30200	41	35	the	the	DET
ajst-30200	41	36	design	design	NOUN
ajst-30200	41	37	parameters	parameter	NOUN
ajst-30200	41	38	of	of	ADP
ajst-30200	41	39	the	the	DET
ajst-30200	41	40	device	device	NOUN
ajst-30200	41	41	,	,	PUNCT
ajst-30200	41	42	such	such	ADJ
ajst-30200	41	43	as	as	ADP
ajst-30200	41	44	the	the	DET
ajst-30200	41	45	size	size	NOUN
ajst-30200	41	46	of	of	ADP
ajst-30200	41	47	the	the	DET
ajst-30200	41	48	water	water	NOUN
ajst-30200	41	49	tank	tank	NOUN
ajst-30200	41	50	,	,	PUNCT
ajst-30200	41	51	turbine	turbine	NOUN
ajst-30200	41	52	design	design	NOUN
ajst-30200	41	53	,	,	PUNCT
ajst-30200	41	54	and	and	CCONJ
ajst-30200	41	55	the	the	DET
ajst-30200	41	56	dimensions	dimension	NOUN
ajst-30200	41	57	of	of	ADP
ajst-30200	41	58	the	the	DET
ajst-30200	41	59	intake	intake	NOUN
ajst-30200	41	60	and	and	CCONJ
ajst-30200	41	61	exhaust	exhaust	NOUN
ajst-30200	41	62	ports	port	NOUN
ajst-30200	41	63	.	.	PUNCT
ajst-30200	42	1	therefore	therefore	ADV
ajst-30200	42	2	,	,	PUNCT
ajst-30200	42	3	parameter	parameter	NOUN
ajst-30200	42	4	optimization	optimization	NOUN
ajst-30200	42	5	is	be	AUX
ajst-30200	42	6	an	an	DET
ajst-30200	42	7	essential	essential	ADJ
ajst-30200	42	8	method	method	NOUN
ajst-30200	42	9	for	for	ADP
ajst-30200	42	10	improving	improve	VERB
ajst-30200	42	11	energy	energy	NOUN
ajst-30200	42	12	conversion	conversion	NOUN
ajst-30200	42	13	efficiency	efficiency	NOUN
ajst-30200	42	14	.	.	PUNCT
ajst-30200	43	1	by	by	ADP
ajst-30200	43	2	optimizing	optimize	VERB
ajst-30200	43	3	control	control	NOUN
ajst-30200	43	4	objectives	objective	NOUN
ajst-30200	43	5	and	and	CCONJ
ajst-30200	43	6	designing	design	VERB
ajst-30200	43	7	operation	operation	NOUN
ajst-30200	43	8	strategies	strategy	NOUN
ajst-30200	43	9	that	that	PRON
ajst-30200	43	10	adapt	adapt	VERB
ajst-30200	43	11	to	to	ADP
ajst-30200	43	12	different	different	ADJ
ajst-30200	43	13	wave	wave	NOUN
ajst-30200	43	14	conditions	condition	NOUN
ajst-30200	43	15	,	,	PUNCT
ajst-30200	43	16	the	the	DET
ajst-30200	43	17	maximum	maximum	ADJ
ajst-30200	43	18	capture	capture	NOUN
ajst-30200	43	19	of	of	ADP
ajst-30200	43	20	wave	wave	NOUN
ajst-30200	43	21	energy	energy	NOUN
ajst-30200	43	22	can	can	AUX
ajst-30200	43	23	be	be	AUX
ajst-30200	43	24	achieved	achieve	VERB
ajst-30200	43	25	.	.	PUNCT
ajst-30200	44	1	control	control	NOUN
ajst-30200	44	2	objectives	objective	NOUN
ajst-30200	44	3	should	should	AUX
ajst-30200	44	4	consider	consider	VERB
ajst-30200	44	5	factors	factor	NOUN
ajst-30200	44	6	such	such	ADJ
ajst-30200	44	7	as	as	ADP
ajst-30200	44	8	energy	energy	NOUN
ajst-30200	44	9	conversion	conversion	NOUN
ajst-30200	44	10	efficiency	efficiency	NOUN
ajst-30200	44	11	,	,	PUNCT
ajst-30200	44	12	system	system	NOUN
ajst-30200	44	13	stability	stability	NOUN
ajst-30200	44	14	,	,	PUNCT
ajst-30200	44	15	and	and	CCONJ
ajst-30200	44	16	long	long	ADJ
ajst-30200	44	17	-	-	PUNCT
ajst-30200	44	18	term	term	NOUN
ajst-30200	44	19	reliability	reliability	NOUN
ajst-30200	44	20	.	.	PUNCT
ajst-30200	45	1	3	3	X
ajst-30200	45	2	.	.	X
ajst-30200	45	3	optimization	optimization	NOUN
ajst-30200	45	4	strategies	strategy	NOUN
ajst-30200	45	5	for	for	ADP
ajst-30200	45	6	improving	improve	VERB
ajst-30200	45	7	energy	energy	NOUN
ajst-30200	45	8	conversion	conversion	NOUN
ajst-30200	45	9	efficiency	efficiency	NOUN
ajst-30200	45	10	optimization	optimization	NOUN
ajst-30200	45	11	strategies	strategy	NOUN
ajst-30200	45	12	for	for	ADP
ajst-30200	45	13	improving	improve	VERB
ajst-30200	45	14	energy	energy	NOUN
ajst-30200	45	15	conversion	conversion	NOUN
ajst-30200	45	16	efficiency	efficiency	NOUN
ajst-30200	45	17	include	include	VERB
ajst-30200	45	18	several	several	ADJ
ajst-30200	45	19	aspects	aspect	NOUN
ajst-30200	45	20	.	.	PUNCT
ajst-30200	46	1	first	first	ADV
ajst-30200	46	2	,	,	PUNCT
ajst-30200	46	3	the	the	DET
ajst-30200	46	4	shape	shape	NOUN
ajst-30200	46	5	of	of	ADP
ajst-30200	46	6	the	the	DET
ajst-30200	46	7	water	water	NOUN
ajst-30200	46	8	tank	tank	NOUN
ajst-30200	46	9	and	and	CCONJ
ajst-30200	46	10	the	the	DET
ajst-30200	46	11	layout	layout	NOUN
ajst-30200	46	12	of	of	ADP
ajst-30200	46	13	the	the	DET
ajst-30200	46	14	turbine	turbine	NOUN
ajst-30200	46	15	should	should	AUX
ajst-30200	46	16	be	be	AUX
ajst-30200	46	17	adjusted	adjust	VERB
ajst-30200	46	18	to	to	PART
ajst-30200	46	19	achieve	achieve	VERB
ajst-30200	46	20	optimal	optimal	ADJ
ajst-30200	46	21	wave	wave	NOUN
ajst-30200	46	22	energy	energy	NOUN
ajst-30200	46	23	capture	capture	NOUN
ajst-30200	46	24	.	.	PUNCT
ajst-30200	47	1	second	second	ADJ
ajst-30200	47	2	,	,	PUNCT
ajst-30200	47	3	adaptive	adaptive	ADJ
ajst-30200	47	4	control	control	NOUN
ajst-30200	47	5	strategies	strategy	NOUN
ajst-30200	47	6	can	can	AUX
ajst-30200	47	7	be	be	AUX
ajst-30200	47	8	employed	employ	VERB
ajst-30200	47	9	to	to	PART
ajst-30200	47	10	dynamically	dynamically	ADV
ajst-30200	47	11	adjust	adjust	VERB
ajst-30200	47	12	the	the	DET
ajst-30200	47	13	operating	operate	VERB
ajst-30200	47	14	state	state	NOUN
ajst-30200	47	15	of	of	ADP
ajst-30200	47	16	the	the	DET
ajst-30200	47	17	turbine	turbine	NOUN
ajst-30200	47	18	,	,	PUNCT
ajst-30200	47	19	optimizing	optimize	VERB
ajst-30200	47	20	airflow	airflow	NOUN
ajst-30200	47	21	and	and	CCONJ
ajst-30200	47	22	pressure	pressure	NOUN
ajst-30200	47	23	based	base	VERB
ajst-30200	47	24	on	on	ADP
ajst-30200	47	25	real	real	ADJ
ajst-30200	47	26	-	-	PUNCT
ajst-30200	47	27	time	time	NOUN
ajst-30200	47	28	changes	change	NOUN
ajst-30200	47	29	in	in	ADP
ajst-30200	47	30	the	the	DET
ajst-30200	47	31	waves	wave	NOUN
ajst-30200	47	32	.	.	PUNCT
ajst-30200	48	1	intelligent	intelligent	ADJ
ajst-30200	48	2	control	control	NOUN
ajst-30200	48	3	methods	method	NOUN
ajst-30200	48	4	,	,	PUNCT
ajst-30200	48	5	such	such	ADJ
ajst-30200	48	6	as	as	ADP
ajst-30200	48	7	deep	deep	ADJ
ajst-30200	48	8	reinforcement	reinforcement	NOUN
ajst-30200	48	9	learning	learning	NOUN
ajst-30200	48	10	,	,	PUNCT
ajst-30200	48	11	can	can	AUX
ajst-30200	48	12	continuously	continuously	ADV
ajst-30200	48	13	adjust	adjust	VERB
ajst-30200	48	14	control	control	NOUN
ajst-30200	48	15	strategies	strategy	NOUN
ajst-30200	48	16	to	to	PART
ajst-30200	48	17	optimize	optimize	VERB
ajst-30200	48	18	the	the	DET
ajst-30200	48	19	performance	performance	NOUN
ajst-30200	48	20	of	of	ADP
ajst-30200	48	21	the	the	DET
ajst-30200	48	22	device	device	NOUN
ajst-30200	48	23	under	under	ADP
ajst-30200	48	24	various	various	ADJ
ajst-30200	48	25	sea	sea	NOUN
ajst-30200	48	26	conditions	condition	NOUN
ajst-30200	48	27	.	.	PUNCT
ajst-30200	49	1	additionally	additionally	ADV
ajst-30200	49	2	,	,	PUNCT
ajst-30200	49	3	given	give	VERB
ajst-30200	49	4	the	the	DET
ajst-30200	49	5	variability	variability	NOUN
ajst-30200	49	6	of	of	ADP
ajst-30200	49	7	wave	wave	NOUN
ajst-30200	49	8	energy	energy	NOUN
ajst-30200	49	9	,	,	PUNCT
ajst-30200	49	10	predictive	predictive	ADJ
ajst-30200	49	11	control	control	NOUN
ajst-30200	49	12	and	and	CCONJ
ajst-30200	49	13	fault	fault	VERB
ajst-30200	49	14	diagnosis	diagnosis	NOUN
ajst-30200	49	15	methods	method	NOUN
ajst-30200	49	16	can	can	AUX
ajst-30200	49	17	be	be	AUX
ajst-30200	49	18	used	use	VERB
ajst-30200	49	19	to	to	PART
ajst-30200	49	20	maintain	maintain	VERB
ajst-30200	49	21	and	and	CCONJ
ajst-30200	49	22	adjust	adjust	VERB
ajst-30200	49	23	the	the	DET
ajst-30200	49	24	system	system	NOUN
ajst-30200	49	25	,	,	PUNCT
ajst-30200	49	26	further	far	ADV
ajst-30200	49	27	enhancing	enhance	VERB
ajst-30200	49	28	efficiency	efficiency	NOUN
ajst-30200	49	29	.	.	PUNCT
ajst-30200	50	1	2	2	X
ajst-30200	50	2	.	.	NUM
ajst-30200	50	3	basic	basic	ADJ
ajst-30200	50	4	principles	principle	NOUN
ajst-30200	50	5	and	and	CCONJ
ajst-30200	50	6	applications	application	NOUN
ajst-30200	50	7	of	of	ADP
ajst-30200	50	8	deep	deep	ADJ
ajst-30200	50	9	reinforcement	reinforcement	NOUN
ajst-30200	50	10	learning	learning	NOUN
ajst-30200	50	11	(	(	PUNCT
ajst-30200	50	12	a	a	X
ajst-30200	50	13	)	)	PUNCT
ajst-30200	50	14	overview	overview	NOUN
ajst-30200	50	15	of	of	ADP
ajst-30200	50	16	deep	deep	ADJ
ajst-30200	50	17	reinforcement	reinforcement	NOUN
ajst-30200	50	18	learning	learning	NOUN
ajst-30200	50	19	1	1	NUM
ajst-30200	50	20	.	.	PUNCT
ajst-30200	50	21	basic	basic	ADJ
ajst-30200	50	22	concepts	concept	NOUN
ajst-30200	50	23	of	of	ADP
ajst-30200	50	24	reinforcement	reinforcement	NOUN
ajst-30200	50	25	learning	learning	NOUN
ajst-30200	50	26	reinforcement	reinforcement	NOUN
ajst-30200	50	27	learning	learning	NOUN
ajst-30200	50	28	(	(	PUNCT
ajst-30200	50	29	rl	rl	NOUN
ajst-30200	50	30	)	)	PUNCT
ajst-30200	50	31	is	be	AUX
ajst-30200	50	32	a	a	DET
ajst-30200	50	33	paradigm	paradigm	NOUN
ajst-30200	50	34	in	in	ADP
ajst-30200	50	35	machine	machine	NOUN
ajst-30200	50	36	learning	learning	NOUN
ajst-30200	50	37	where	where	SCONJ
ajst-30200	50	38	an	an	DET
ajst-30200	50	39	agent	agent	NOUN
ajst-30200	50	40	learns	learn	VERB
ajst-30200	50	41	decision	decision	NOUN
ajst-30200	50	42	-	-	PUNCT
ajst-30200	50	43	making	making	NOUN
ajst-30200	50	44	strategies	strategy	NOUN
ajst-30200	50	45	through	through	ADP
ajst-30200	50	46	interactions	interaction	NOUN
ajst-30200	50	47	with	with	ADP
ajst-30200	50	48	an	an	DET
ajst-30200	50	49	environment	environment	NOUN
ajst-30200	50	50	.	.	PUNCT
ajst-30200	51	1	the	the	DET
ajst-30200	51	2	agent	agent	NOUN
ajst-30200	51	3	takes	take	VERB
ajst-30200	51	4	actions	action	NOUN
ajst-30200	51	5	in	in	ADP
ajst-30200	51	6	the	the	DET
ajst-30200	51	7	environment	environment	NOUN
ajst-30200	51	8	and	and	CCONJ
ajst-30200	51	9	adjusts	adjust	VERB
ajst-30200	51	10	its	its	PRON
ajst-30200	51	11	strategy	strategy	NOUN
ajst-30200	51	12	based	base	VERB
ajst-30200	51	13	on	on	ADP
ajst-30200	51	14	the	the	DET
ajst-30200	51	15	feedback	feedback	NOUN
ajst-30200	51	16	(	(	PUNCT
ajst-30200	51	17	rewards	reward	NOUN
ajst-30200	51	18	or	or	CCONJ
ajst-30200	51	19	punishments	punishment	NOUN
ajst-30200	51	20	)	)	PUNCT
ajst-30200	51	21	from	from	ADP
ajst-30200	51	22	the	the	DET
ajst-30200	51	23	environment	environment	NOUN
ajst-30200	51	24	.	.	PUNCT
ajst-30200	52	1	the	the	DET
ajst-30200	52	2	core	core	NOUN
ajst-30200	52	3	of	of	ADP
ajst-30200	52	4	reinforcement	reinforcement	NOUN
ajst-30200	52	5	learning	learning	NOUN
ajst-30200	52	6	is	be	AUX
ajst-30200	52	7	to	to	PART
ajst-30200	52	8	find	find	VERB
ajst-30200	52	9	the	the	DET
ajst-30200	52	10	optimal	optimal	ADJ
ajst-30200	52	11	behavior	behavior	NOUN
ajst-30200	52	12	strategy	strategy	NOUN
ajst-30200	52	13	by	by	ADP
ajst-30200	52	14	maximizing	maximize	VERB
ajst-30200	52	15	cumulative	cumulative	ADJ
ajst-30200	52	16	rewards	reward	NOUN
ajst-30200	52	17	.	.	PUNCT
ajst-30200	53	1	unlike	unlike	ADP
ajst-30200	53	2	supervised	supervised	ADJ
ajst-30200	53	3	learning	learning	NOUN
ajst-30200	53	4	,	,	PUNCT
ajst-30200	53	5	rl	rl	PROPN
ajst-30200	53	6	does	do	AUX
ajst-30200	53	7	not	not	PART
ajst-30200	53	8	rely	rely	VERB
ajst-30200	53	9	on	on	ADP
ajst-30200	53	10	pre	pre	ADJ
ajst-30200	53	11	-	-	ADJ
ajst-30200	53	12	labeled	labeled	ADJ
ajst-30200	53	13	data	datum	NOUN
ajst-30200	53	14	but	but	CCONJ
ajst-30200	53	15	learns	learn	VERB
ajst-30200	53	16	the	the	DET
ajst-30200	53	17	best	good	ADJ
ajst-30200	53	18	strategy	strategy	NOUN
ajst-30200	53	19	through	through	ADP
ajst-30200	53	20	exploration	exploration	NOUN
ajst-30200	53	21	and	and	CCONJ
ajst-30200	53	22	trial	trial	NOUN
ajst-30200	53	23	-	-	PUNCT
ajst-30200	53	24	and	and	CCONJ
ajst-30200	53	25	-	-	PUNCT
ajst-30200	53	26	error	error	NOUN
ajst-30200	53	27	.	.	PUNCT
ajst-30200	54	1	in	in	ADP
ajst-30200	54	2	the	the	DET
ajst-30200	54	3	framework	framework	NOUN
ajst-30200	54	4	of	of	ADP
ajst-30200	54	5	reinforcement	reinforcement	NOUN
ajst-30200	54	6	learning	learning	NOUN
ajst-30200	54	7	,	,	PUNCT
ajst-30200	54	8	the	the	DET
ajst-30200	54	9	agent	agent	NOUN
ajst-30200	54	10	’s	’s	PART
ajst-30200	54	11	goal	goal	NOUN
ajst-30200	54	12	is	be	AUX
ajst-30200	54	13	to	to	PART
ajst-30200	54	14	learn	learn	VERB
ajst-30200	54	15	an	an	DET
ajst-30200	54	16	optimal	optimal	ADJ
ajst-30200	54	17	strategy	strategy	NOUN
ajst-30200	54	18	such	such	ADJ
ajst-30200	54	19	that	that	SCONJ
ajst-30200	54	20	performing	perform	VERB
ajst-30200	54	21	a	a	DET
ajst-30200	54	22	particular	particular	ADJ
ajst-30200	54	23	action	action	NOUN
ajst-30200	54	24	in	in	ADP
ajst-30200	54	25	a	a	DET
ajst-30200	54	26	given	give	VERB
ajst-30200	54	27	state	state	NOUN
ajst-30200	54	28	will	will	AUX
ajst-30200	54	29	yield	yield	VERB
ajst-30200	54	30	the	the	DET
ajst-30200	54	31	maximum	maximum	ADJ
ajst-30200	54	32	longterm	longterm	PROPN
ajst-30200	54	33	reward	reward	NOUN
ajst-30200	54	34	.	.	PUNCT
ajst-30200	55	1	through	through	ADP
ajst-30200	55	2	interactions	interaction	NOUN
ajst-30200	55	3	with	with	ADP
ajst-30200	55	4	the	the	DET
ajst-30200	55	5	environment	environment	NOUN
ajst-30200	55	6	,	,	PUNCT
ajst-30200	55	7	the	the	DET
ajst-30200	55	8	agent	agent	NOUN
ajst-30200	55	9	continually	continually	ADV
ajst-30200	55	10	evaluates	evaluate	VERB
ajst-30200	55	11	the	the	DET
ajst-30200	55	12	outcomes	outcome	NOUN
ajst-30200	55	13	of	of	ADP
ajst-30200	55	14	its	its	PRON
ajst-30200	55	15	actions	action	NOUN
ajst-30200	55	16	,	,	PUNCT
ajst-30200	55	17	adjusting	adjust	VERB
ajst-30200	55	18	its	its	PRON
ajst-30200	55	19	future	future	ADJ
ajst-30200	55	20	decisions	decision	NOUN
ajst-30200	55	21	to	to	PART
ajst-30200	55	22	gradually	gradually	ADV
ajst-30200	55	23	approach	approach	VERB
ajst-30200	55	24	the	the	DET
ajst-30200	55	25	optimal	optimal	ADJ
ajst-30200	55	26	strategy.[3	strategy.[3	NOUN
ajst-30200	55	27	]	]	SYM
ajst-30200	55	28	.	.	PUNCT
ajst-30200	56	1	2	2	X
ajst-30200	56	2	.	.	X
ajst-30200	56	3	advantages	advantage	NOUN
ajst-30200	56	4	of	of	ADP
ajst-30200	56	5	combining	combine	VERB
ajst-30200	56	6	deep	deep	ADJ
ajst-30200	56	7	learning	learning	NOUN
ajst-30200	56	8	with	with	ADP
ajst-30200	56	9	reinforcement	reinforcement	NOUN
ajst-30200	56	10	learning	learning	NOUN
ajst-30200	56	11	deep	deep	ADJ
ajst-30200	56	12	learning	learning	NOUN
ajst-30200	56	13	(	(	PUNCT
ajst-30200	56	14	dl	dl	INTJ
ajst-30200	56	15	)	)	PUNCT
ajst-30200	56	16	is	be	AUX
ajst-30200	56	17	a	a	DET
ajst-30200	56	18	learning	learning	NOUN
ajst-30200	56	19	method	method	NOUN
ajst-30200	56	20	based	base	VERB
ajst-30200	56	21	on	on	ADP
ajst-30200	56	22	deep	deep	ADJ
ajst-30200	56	23	neural	neural	ADJ
ajst-30200	56	24	networks	network	NOUN
ajst-30200	56	25	that	that	PRON
ajst-30200	56	26	can	can	AUX
ajst-30200	56	27	automatically	automatically	ADV
ajst-30200	56	28	extract	extract	VERB
ajst-30200	56	29	features	feature	NOUN
ajst-30200	56	30	from	from	ADP
ajst-30200	56	31	data	datum	NOUN
ajst-30200	56	32	through	through	ADP
ajst-30200	56	33	multi	multi	ADJ
ajst-30200	56	34	-	-	ADJ
ajst-30200	56	35	layer	layer	ADJ
ajst-30200	56	36	structures	structure	NOUN
ajst-30200	56	37	.	.	PUNCT
ajst-30200	57	1	in	in	ADP
ajst-30200	57	2	reinforcement	reinforcement	NOUN
ajst-30200	57	3	learning	learning	NOUN
ajst-30200	57	4	,	,	PUNCT
ajst-30200	57	5	traditional	traditional	ADJ
ajst-30200	57	6	rl	rl	NOUN
ajst-30200	57	7	methods	method	NOUN
ajst-30200	57	8	(	(	PUNCT
ajst-30200	57	9	such	such	ADJ
ajst-30200	57	10	as	as	ADP
ajst-30200	57	11	q	q	NOUN
ajst-30200	57	12	-	-	PUNCT
ajst-30200	57	13	learning	learning	NOUN
ajst-30200	57	14	)	)	PUNCT
ajst-30200	57	15	are	be	AUX
ajst-30200	57	16	limited	limit	VERB
ajst-30200	57	17	by	by	ADP
ajst-30200	57	18	the	the	DET
ajst-30200	57	19	scale	scale	NOUN
ajst-30200	57	20	of	of	ADP
ajst-30200	57	21	state	state	NOUN
ajst-30200	57	22	and	and	CCONJ
ajst-30200	57	23	action	action	NOUN
ajst-30200	57	24	spaces	space	NOUN
ajst-30200	57	25	,	,	PUNCT
ajst-30200	57	26	especially	especially	ADV
ajst-30200	57	27	in	in	ADP
ajst-30200	57	28	complex	complex	ADJ
ajst-30200	57	29	,	,	PUNCT
ajst-30200	57	30	highdimensional	highdimensional	ADJ
ajst-30200	57	31	problems	problem	NOUN
ajst-30200	57	32	where	where	SCONJ
ajst-30200	57	33	traditional	traditional	ADJ
ajst-30200	57	34	methods	method	NOUN
ajst-30200	57	35	are	be	AUX
ajst-30200	57	36	less	less	ADV
ajst-30200	57	37	efficient	efficient	ADJ
ajst-30200	57	38	.	.	PUNCT
ajst-30200	58	1	by	by	ADP
ajst-30200	58	2	combining	combine	VERB
ajst-30200	58	3	deep	deep	ADJ
ajst-30200	58	4	learning	learning	NOUN
ajst-30200	58	5	,	,	PUNCT
ajst-30200	58	6	deep	deep	ADJ
ajst-30200	58	7	reinforcement	reinforcement	NOUN
ajst-30200	58	8	learning	learning	NOUN
ajst-30200	58	9	(	(	PUNCT
ajst-30200	58	10	drl	drl	PROPN
ajst-30200	58	11	)	)	PUNCT
ajst-30200	58	12	can	can	AUX
ajst-30200	58	13	handle	handle	VERB
ajst-30200	58	14	larger	large	ADJ
ajst-30200	58	15	and	and	CCONJ
ajst-30200	58	16	more	more	ADV
ajst-30200	58	17	complex	complex	ADJ
ajst-30200	58	18	state	state	NOUN
ajst-30200	58	19	and	and	CCONJ
ajst-30200	58	20	action	action	NOUN
ajst-30200	58	21	spaces	space	NOUN
ajst-30200	58	22	,	,	PUNCT
ajst-30200	58	23	significantly	significantly	ADV
ajst-30200	58	24	enhancing	enhance	VERB
ajst-30200	58	25	the	the	DET
ajst-30200	58	26	agent	agent	NOUN
ajst-30200	58	27	's	's	PART
ajst-30200	58	28	learning	learn	VERB
ajst-30200	58	29	capabilities	capability	NOUN
ajst-30200	58	30	and	and	CCONJ
ajst-30200	58	31	performance	performance	NOUN
ajst-30200	58	32	.	.	PUNCT
ajst-30200	59	1	the	the	DET
ajst-30200	59	2	introduction	introduction	NOUN
ajst-30200	59	3	of	of	ADP
ajst-30200	59	4	deep	deep	ADJ
ajst-30200	59	5	learning	learning	NOUN
ajst-30200	59	6	allows	allow	VERB
ajst-30200	59	7	reinforcement	reinforcement	NOUN
ajst-30200	59	8	learning	learning	NOUN
ajst-30200	59	9	to	to	PART
ajst-30200	59	10	automatically	automatically	ADV
ajst-30200	59	11	extract	extract	VERB
ajst-30200	59	12	features	feature	NOUN
ajst-30200	59	13	from	from	ADP
ajst-30200	59	14	raw	raw	ADJ
ajst-30200	59	15	data	datum	NOUN
ajst-30200	59	16	without	without	ADP
ajst-30200	59	17	the	the	DET
ajst-30200	59	18	need	need	NOUN
ajst-30200	59	19	for	for	ADP
ajst-30200	59	20	manual	manual	ADJ
ajst-30200	59	21	feature	feature	NOUN
ajst-30200	59	22	design	design	NOUN
ajst-30200	59	23	.	.	PUNCT
ajst-30200	60	1	with	with	ADP
ajst-30200	60	2	multi	multi	ADJ
ajst-30200	60	3	-	-	ADJ
ajst-30200	60	4	layer	layer	NOUN
ajst-30200	60	5	processing	processing	NOUN
ajst-30200	60	6	through	through	ADP
ajst-30200	60	7	deep	deep	ADJ
ajst-30200	60	8	neural	neural	ADJ
ajst-30200	60	9	networks	network	NOUN
ajst-30200	60	10	,	,	PUNCT
ajst-30200	60	11	drl	drl	PROPN
ajst-30200	60	12	can	can	AUX
ajst-30200	60	13	execute	execute	VERB
ajst-30200	60	14	complex	complex	ADJ
ajst-30200	60	15	decision	decision	NOUN
ajst-30200	60	16	tasks	task	NOUN
ajst-30200	60	17	in	in	ADP
ajst-30200	60	18	complex	complex	ADJ
ajst-30200	60	19	environments	environment	NOUN
ajst-30200	60	20	,	,	PUNCT
ajst-30200	60	21	such	such	ADJ
ajst-30200	60	22	as	as	ADP
ajst-30200	60	23	image	image	NOUN
ajst-30200	60	24	processing	processing	NOUN
ajst-30200	60	25	,	,	PUNCT
ajst-30200	60	26	natural	natural	ADJ
ajst-30200	60	27	language	language	NOUN
ajst-30200	60	28	understanding	understanding	NOUN
ajst-30200	60	29	,	,	PUNCT
ajst-30200	60	30	and	and	CCONJ
ajst-30200	60	31	complex	complex	ADJ
ajst-30200	60	32	control	control	NOUN
ajst-30200	60	33	tasks	task	NOUN
ajst-30200	60	34	.	.	PUNCT
ajst-30200	61	1	therefore	therefore	ADV
ajst-30200	61	2	,	,	PUNCT
ajst-30200	61	3	drl	drl	PROPN
ajst-30200	61	4	has	have	VERB
ajst-30200	61	5	significant	significant	ADJ
ajst-30200	61	6	advantages	advantage	NOUN
ajst-30200	61	7	in	in	ADP
ajst-30200	61	8	many	many	ADJ
ajst-30200	61	9	fields	field	NOUN
ajst-30200	61	10	,	,	PUNCT
ajst-30200	61	11	especially	especially	ADV
ajst-30200	61	12	in	in	ADP
ajst-30200	61	13	high	high	ADJ
ajst-30200	61	14	-	-	PUNCT
ajst-30200	61	15	dimensional	dimensional	ADJ
ajst-30200	61	16	and	and	CCONJ
ajst-30200	61	17	complex	complex	ADJ
ajst-30200	61	18	decision	decision	NOUN
ajst-30200	61	19	-	-	PUNCT
ajst-30200	61	20	making	make	VERB
ajst-30200	61	21	problems	problem	NOUN
ajst-30200	61	22	.	.	PUNCT
ajst-30200	62	1	3	3	X
ajst-30200	62	2	.	.	X
ajst-30200	62	3	core	core	NOUN
ajst-30200	62	4	algorithms	algorithm	NOUN
ajst-30200	62	5	of	of	ADP
ajst-30200	62	6	deep	deep	ADJ
ajst-30200	62	7	reinforcement	reinforcement	NOUN
ajst-30200	62	8	learning	learn	VERB
ajst-30200	62	9	deep	deep	ADJ
ajst-30200	62	10	reinforcement	reinforcement	NOUN
ajst-30200	62	11	learning	learning	NOUN
ajst-30200	62	12	combines	combine	VERB
ajst-30200	62	13	the	the	DET
ajst-30200	62	14	decisionmaking	decisionmake	VERB
ajst-30200	62	15	framework	framework	NOUN
ajst-30200	62	16	of	of	ADP
ajst-30200	62	17	reinforcement	reinforcement	NOUN
ajst-30200	62	18	learning	learning	NOUN
ajst-30200	62	19	with	with	ADP
ajst-30200	62	20	the	the	DET
ajst-30200	62	21	feature	feature	NOUN
ajst-30200	62	22	extraction	extraction	NOUN
ajst-30200	62	23	capabilities	capability	NOUN
ajst-30200	62	24	of	of	ADP
ajst-30200	62	25	deep	deep	ADJ
ajst-30200	62	26	learning	learning	NOUN
ajst-30200	62	27	.	.	PUNCT
ajst-30200	63	1	the	the	DET
ajst-30200	63	2	core	core	ADJ
ajst-30200	63	3	algorithms	algorithm	NOUN
ajst-30200	63	4	mainly	mainly	ADV
ajst-30200	63	5	include	include	VERB
ajst-30200	63	6	the	the	DET
ajst-30200	63	7	following	follow	VERB
ajst-30200	63	8	:	:	PUNCT
ajst-30200	63	9	deep	deep	ADJ
ajst-30200	63	10	q	q	NOUN
ajst-30200	63	11	-	-	PUNCT
ajst-30200	63	12	network	network	NOUN
ajst-30200	63	13	(	(	PUNCT
ajst-30200	63	14	dqn	dqn	PROPN
ajst-30200	63	15	):	):	PUNCT
ajst-30200	63	16	dqn	dqn	NOUN
ajst-30200	63	17	is	be	AUX
ajst-30200	63	18	an	an	DET
ajst-30200	63	19	important	important	ADJ
ajst-30200	63	20	breakthrough	breakthrough	NOUN
ajst-30200	63	21	in	in	ADP
ajst-30200	63	22	reinforcement	reinforcement	NOUN
ajst-30200	63	23	learning	learning	NOUN
ajst-30200	63	24	.	.	PUNCT
ajst-30200	64	1	it	it	PRON
ajst-30200	64	2	uses	use	VERB
ajst-30200	64	3	deep	deep	ADJ
ajst-30200	64	4	neural	neural	ADJ
ajst-30200	64	5	networks	network	NOUN
ajst-30200	64	6	to	to	PART
ajst-30200	64	7	approximate	approximate	VERB
ajst-30200	64	8	the	the	DET
ajst-30200	64	9	q	q	ADJ
ajst-30200	64	10	-	-	PUNCT
ajst-30200	64	11	value	value	NOUN
ajst-30200	64	12	function	function	NOUN
ajst-30200	64	13	,	,	PUNCT
ajst-30200	64	14	solving	solve	VERB
ajst-30200	64	15	the	the	DET
ajst-30200	64	16	bottleneck	bottleneck	NOUN
ajst-30200	64	17	of	of	ADP
ajst-30200	64	18	traditional	traditional	ADJ
ajst-30200	64	19	q	q	NOUN
ajst-30200	64	20	-	-	PUNCT
ajst-30200	64	21	learning	learning	NOUN
ajst-30200	64	22	in	in	ADP
ajst-30200	64	23	high	high	ADJ
ajst-30200	64	24	-	-	PUNCT
ajst-30200	64	25	dimensional	dimensional	ADJ
ajst-30200	64	26	state	state	NOUN
ajst-30200	64	27	spaces	space	NOUN
ajst-30200	64	28	.	.	PUNCT
ajst-30200	65	1	dqn	dqn	NOUN
ajst-30200	65	2	enhances	enhance	VERB
ajst-30200	65	3	the	the	DET
ajst-30200	65	4	training	training	NOUN
ajst-30200	65	5	process	process	NOUN
ajst-30200	65	6	by	by	ADP
ajst-30200	65	7	using	use	VERB
ajst-30200	65	8	experience	experience	NOUN
ajst-30200	65	9	replay	replay	NOUN
ajst-30200	65	10	and	and	CCONJ
ajst-30200	65	11	target	target	NOUN
ajst-30200	65	12	networks	network	NOUN
ajst-30200	65	13	,	,	PUNCT
ajst-30200	65	14	greatly	greatly	ADV
ajst-30200	65	15	improving	improve	VERB
ajst-30200	65	16	qlearning	qlearning	NOUN
ajst-30200	65	17	’s	’s	PART
ajst-30200	65	18	application	application	NOUN
ajst-30200	65	19	in	in	ADP
ajst-30200	65	20	complex	complex	ADJ
ajst-30200	65	21	problems	problem	NOUN
ajst-30200	65	22	.	.	PUNCT
ajst-30200	66	1	policy	policy	NOUN
ajst-30200	66	2	gradient	gradient	NOUN
ajst-30200	66	3	methods	method	NOUN
ajst-30200	66	4	:	:	PUNCT
ajst-30200	66	5	policy	policy	NOUN
ajst-30200	66	6	gradient	gradient	NOUN
ajst-30200	66	7	methods	method	NOUN
ajst-30200	66	8	directly	directly	ADV
ajst-30200	66	9	optimize	optimize	VERB
ajst-30200	66	10	the	the	DET
ajst-30200	66	11	agent	agent	NOUN
ajst-30200	66	12	's	's	PART
ajst-30200	66	13	behavior	behavior	NOUN
ajst-30200	66	14	strategy	strategy	NOUN
ajst-30200	66	15	rather	rather	ADV
ajst-30200	66	16	than	than	ADP
ajst-30200	66	17	indirectly	indirectly	ADV
ajst-30200	66	18	optimizing	optimize	VERB
ajst-30200	66	19	the	the	DET
ajst-30200	66	20	value	value	NOUN
ajst-30200	66	21	function	function	NOUN
ajst-30200	66	22	.	.	PUNCT
ajst-30200	67	1	the	the	DET
ajst-30200	67	2	policy	policy	NOUN
ajst-30200	67	3	gradient	gradient	NOUN
ajst-30200	67	4	algorithm	algorithm	NOUN
ajst-30200	67	5	calculates	calculate	VERB
ajst-30200	67	6	the	the	DET
ajst-30200	67	7	gradient	gradient	NOUN
ajst-30200	67	8	of	of	ADP
ajst-30200	67	9	the	the	DET
ajst-30200	67	10	policy	policy	NOUN
ajst-30200	67	11	and	and	CCONJ
ajst-30200	67	12	updates	update	VERB
ajst-30200	67	13	the	the	DET
ajst-30200	67	14	policy	policy	NOUN
ajst-30200	67	15	parameters	parameter	NOUN
ajst-30200	67	16	to	to	PART
ajst-30200	67	17	improve	improve	VERB
ajst-30200	67	18	performance	performance	NOUN
ajst-30200	67	19	.	.	PUNCT
ajst-30200	68	1	common	common	ADJ
ajst-30200	68	2	policy	policy	NOUN
ajst-30200	68	3	gradient	gradient	NOUN
ajst-30200	68	4	methods	method	NOUN
ajst-30200	68	5	include	include	VERB
ajst-30200	68	6	the	the	DET
ajst-30200	68	7	reinforce	reinforce	NOUN
ajst-30200	68	8	algorithm	algorithm	NOUN
ajst-30200	68	9	.	.	PUNCT
ajst-30200	69	1	actor	actor	NOUN
ajst-30200	69	2	-	-	PUNCT
ajst-30200	69	3	critic	critic	NOUN
ajst-30200	69	4	methods	method	NOUN
ajst-30200	69	5	:	:	PUNCT
ajst-30200	69	6	actor	actor	NOUN
ajst-30200	69	7	-	-	PUNCT
ajst-30200	69	8	critic	critic	NOUN
ajst-30200	69	9	methods	method	NOUN
ajst-30200	69	10	combine	combine	VERB
ajst-30200	69	11	value	value	NOUN
ajst-30200	69	12	function	function	NOUN
ajst-30200	69	13	methods	method	NOUN
ajst-30200	69	14	and	and	CCONJ
ajst-30200	69	15	policy	policy	NOUN
ajst-30200	69	16	methods	method	NOUN
ajst-30200	69	17	by	by	ADP
ajst-30200	69	18	simultaneously	simultaneously	ADV
ajst-30200	69	19	learning	learn	VERB
ajst-30200	69	20	a	a	DET
ajst-30200	69	21	policy	policy	NOUN
ajst-30200	69	22	(	(	PUNCT
ajst-30200	69	23	actor	actor	NOUN
ajst-30200	69	24	)	)	PUNCT
ajst-30200	69	25	and	and	CCONJ
ajst-30200	69	26	a	a	DET
ajst-30200	69	27	value	value	NOUN
ajst-30200	69	28	function	function	NOUN
ajst-30200	69	29	(	(	PUNCT
ajst-30200	69	30	critic	critic	NOUN
ajst-30200	69	31	)	)	PUNCT
ajst-30200	69	32	to	to	PART
ajst-30200	69	33	improve	improve	VERB
ajst-30200	69	34	learning	learn	VERB
ajst-30200	69	35	efficiency	efficiency	NOUN
ajst-30200	69	36	and	and	CCONJ
ajst-30200	69	37	stability	stability	NOUN
ajst-30200	69	38	.	.	PUNCT
ajst-30200	70	1	the	the	DET
ajst-30200	70	2	critic	critic	NOUN
ajst-30200	70	3	evaluates	evaluate	VERB
ajst-30200	70	4	the	the	DET
ajst-30200	70	5	current	current	ADJ
ajst-30200	70	6	policy	policy	NOUN
ajst-30200	70	7	,	,	PUNCT
ajst-30200	70	8	and	and	CCONJ
ajst-30200	70	9	the	the	DET
ajst-30200	70	10	actor	actor	NOUN
ajst-30200	70	11	improves	improve	VERB
ajst-30200	70	12	the	the	DET
ajst-30200	70	13	policy	policy	NOUN
ajst-30200	70	14	based	base	VERB
ajst-30200	70	15	on	on	ADP
ajst-30200	70	16	the	the	DET
ajst-30200	70	17	critic	critic	NOUN
ajst-30200	70	18	’s	’s	PART
ajst-30200	70	19	evaluations	evaluation	NOUN
ajst-30200	70	20	.	.	PUNCT
ajst-30200	71	1	(	(	PUNCT
ajst-30200	71	2	b	b	X
ajst-30200	71	3	)	)	PUNCT
ajst-30200	71	4	introduction	introduction	NOUN
ajst-30200	71	5	to	to	ADP
ajst-30200	71	6	deep	deep	ADJ
ajst-30200	71	7	reinforcement	reinforcement	NOUN
ajst-30200	71	8	learning	learning	NOUN
ajst-30200	71	9	algorithms	algorithms	NOUN
ajst-30200	71	10	1	1	NUM
ajst-30200	71	11	.	.	PUNCT
ajst-30200	72	1	q	q	X
ajst-30200	72	2	-	-	PUNCT
ajst-30200	72	3	learning	learn	VERB
ajst-30200	72	4	and	and	CCONJ
ajst-30200	72	5	deep	deep	ADJ
ajst-30200	72	6	q	q	NOUN
ajst-30200	72	7	-	-	PUNCT
ajst-30200	72	8	network	network	NOUN
ajst-30200	72	9	(	(	PUNCT
ajst-30200	72	10	dqn	dqn	PROPN
ajst-30200	72	11	)	)	PUNCT
ajst-30200	72	12	q	q	NOUN
ajst-30200	72	13	-	-	PUNCT
ajst-30200	72	14	learning	learning	NOUN
ajst-30200	72	15	is	be	AUX
ajst-30200	72	16	a	a	DET
ajst-30200	72	17	value	value	NOUN
ajst-30200	72	18	iteration	iteration	NOUN
ajst-30200	72	19	algorithm	algorithm	NOUN
ajst-30200	72	20	used	use	VERB
ajst-30200	72	21	to	to	PART
ajst-30200	72	22	solve	solve	VERB
ajst-30200	72	23	the	the	DET
ajst-30200	72	24	optimal	optimal	ADJ
ajst-30200	72	25	policy	policy	NOUN
ajst-30200	72	26	problem	problem	NOUN
ajst-30200	72	27	in	in	ADP
ajst-30200	72	28	a	a	DET
ajst-30200	72	29	markov	markov	NOUN
ajst-30200	72	30	decision	decision	NOUN
ajst-30200	72	31	process	process	NOUN
ajst-30200	72	32	(	(	PUNCT
ajst-30200	72	33	mdp	mdp	NOUN
ajst-30200	72	34	)	)	PUNCT
ajst-30200	72	35	.	.	PUNCT
ajst-30200	73	1	in	in	ADP
ajst-30200	73	2	q	q	NOUN
ajst-30200	73	3	-	-	PUNCT
ajst-30200	73	4	learning	learning	NOUN
ajst-30200	73	5	,	,	PUNCT
ajst-30200	73	6	the	the	DET
ajst-30200	73	7	agent	agent	NOUN
ajst-30200	73	8	maintains	maintain	VERB
ajst-30200	73	9	a	a	DET
ajst-30200	73	10	q	q	ADJ
ajst-30200	73	11	-	-	PUNCT
ajst-30200	73	12	value	value	NOUN
ajst-30200	73	13	table	table	NOUN
ajst-30200	73	14	,	,	PUNCT
ajst-30200	73	15	where	where	SCONJ
ajst-30200	73	16	the	the	DET
ajst-30200	73	17	q	q	NOUN
ajst-30200	73	18	-	-	PUNCT
ajst-30200	73	19	value	value	NOUN
ajst-30200	73	20	represents	represent	VERB
ajst-30200	73	21	the	the	DET
ajst-30200	73	22	expected	expect	VERB
ajst-30200	73	23	reward	reward	NOUN
ajst-30200	73	24	for	for	ADP
ajst-30200	73	25	taking	take	VERB
ajst-30200	73	26	a	a	DET
ajst-30200	73	27	particular	particular	ADJ
ajst-30200	73	28	action	action	NOUN
ajst-30200	73	29	in	in	ADP
ajst-30200	73	30	a	a	DET
ajst-30200	73	31	given	give	VERB
ajst-30200	73	32	state	state	NOUN
ajst-30200	73	33	.	.	PUNCT
ajst-30200	74	1	by	by	ADP
ajst-30200	74	2	updating	update	VERB
ajst-30200	74	3	the	the	DET
ajst-30200	74	4	q	q	ADJ
ajst-30200	74	5	-	-	PUNCT
ajst-30200	74	6	value	value	NOUN
ajst-30200	74	7	table	table	NOUN
ajst-30200	74	8	,	,	PUNCT
ajst-30200	74	9	the	the	DET
ajst-30200	74	10	200	200	NUM
ajst-30200	74	11	agent	agent	NOUN
ajst-30200	74	12	can	can	AUX
ajst-30200	74	13	learn	learn	VERB
ajst-30200	74	14	the	the	DET
ajst-30200	74	15	optimal	optimal	ADJ
ajst-30200	74	16	policy	policy	NOUN
ajst-30200	74	17	.	.	PUNCT
ajst-30200	75	1	however	however	ADV
ajst-30200	75	2	,	,	PUNCT
ajst-30200	75	3	q	q	ADJ
ajst-30200	75	4	-	-	PUNCT
ajst-30200	75	5	learning	learn	VERB
ajst-30200	75	6	faces	face	NOUN
ajst-30200	75	7	challenges	challenge	NOUN
ajst-30200	75	8	as	as	SCONJ
ajst-30200	75	9	the	the	DET
ajst-30200	75	10	dimensions	dimension	NOUN
ajst-30200	75	11	of	of	ADP
ajst-30200	75	12	the	the	DET
ajst-30200	75	13	state	state	NOUN
ajst-30200	75	14	and	and	CCONJ
ajst-30200	75	15	action	action	NOUN
ajst-30200	75	16	spaces	space	NOUN
ajst-30200	75	17	grow	grow	VERB
ajst-30200	75	18	,	,	PUNCT
ajst-30200	75	19	making	make	VERB
ajst-30200	75	20	it	it	PRON
ajst-30200	75	21	difficult	difficult	ADJ
ajst-30200	75	22	to	to	PART
ajst-30200	75	23	manage	manage	VERB
ajst-30200	75	24	the	the	DET
ajst-30200	75	25	q	q	ADJ
ajst-30200	75	26	-	-	PUNCT
ajst-30200	75	27	value	value	NOUN
ajst-30200	75	28	table	table	NOUN
ajst-30200	75	29	,	,	PUNCT
ajst-30200	75	30	especially	especially	ADV
ajst-30200	75	31	for	for	ADP
ajst-30200	75	32	continuous	continuous	ADJ
ajst-30200	75	33	state	state	NOUN
ajst-30200	75	34	spaces	space	NOUN
ajst-30200	75	35	.	.	PUNCT
ajst-30200	76	1	to	to	PART
ajst-30200	76	2	address	address	VERB
ajst-30200	76	3	this	this	DET
ajst-30200	76	4	issue	issue	NOUN
ajst-30200	76	5	,	,	PUNCT
ajst-30200	76	6	the	the	DET
ajst-30200	76	7	deep	deep	ADJ
ajst-30200	76	8	q	q	NOUN
ajst-30200	76	9	-	-	PUNCT
ajst-30200	76	10	network	network	NOUN
ajst-30200	76	11	(	(	PUNCT
ajst-30200	76	12	dqn	dqn	PROPN
ajst-30200	76	13	)	)	PUNCT
ajst-30200	76	14	was	be	AUX
ajst-30200	76	15	developed	develop	VERB
ajst-30200	76	16	.	.	PUNCT
ajst-30200	77	1	dqn	dqn	NOUN
ajst-30200	77	2	uses	use	VERB
ajst-30200	77	3	deep	deep	ADJ
ajst-30200	77	4	neural	neural	ADJ
ajst-30200	77	5	networks	network	NOUN
ajst-30200	77	6	to	to	PART
ajst-30200	77	7	approximate	approximate	VERB
ajst-30200	77	8	the	the	DET
ajst-30200	77	9	q	q	ADJ
ajst-30200	77	10	-	-	PUNCT
ajst-30200	77	11	value	value	NOUN
ajst-30200	77	12	function	function	NOUN
ajst-30200	77	13	,	,	PUNCT
ajst-30200	77	14	enabling	enable	VERB
ajst-30200	77	15	it	it	PRON
ajst-30200	77	16	to	to	PART
ajst-30200	77	17	handle	handle	VERB
ajst-30200	77	18	high	high	ADJ
ajst-30200	77	19	-	-	PUNCT
ajst-30200	77	20	dimensional	dimensional	ADJ
ajst-30200	77	21	state	state	NOUN
ajst-30200	77	22	spaces	space	NOUN
ajst-30200	77	23	.	.	PUNCT
ajst-30200	78	1	the	the	DET
ajst-30200	78	2	core	core	NOUN
ajst-30200	78	3	innovation	innovation	NOUN
ajst-30200	78	4	of	of	ADP
ajst-30200	78	5	dqn	dqn	ADJ
ajst-30200	78	6	lies	lie	NOUN
ajst-30200	78	7	in	in	ADP
ajst-30200	78	8	the	the	DET
ajst-30200	78	9	use	use	NOUN
ajst-30200	78	10	of	of	ADP
ajst-30200	78	11	experience	experience	NOUN
ajst-30200	78	12	replay	replay	NOUN
ajst-30200	78	13	and	and	CCONJ
ajst-30200	78	14	target	target	NOUN
ajst-30200	78	15	networks	network	NOUN
ajst-30200	78	16	to	to	PART
ajst-30200	78	17	stabilize	stabilize	VERB
ajst-30200	78	18	the	the	DET
ajst-30200	78	19	training	training	NOUN
ajst-30200	78	20	process	process	NOUN
ajst-30200	78	21	,	,	PUNCT
ajst-30200	78	22	allowing	allow	VERB
ajst-30200	78	23	dqn	dqn	NOUN
ajst-30200	78	24	to	to	PART
ajst-30200	78	25	make	make	VERB
ajst-30200	78	26	significant	significant	ADJ
ajst-30200	78	27	progress	progress	NOUN
ajst-30200	78	28	in	in	ADP
ajst-30200	78	29	complex	complex	ADJ
ajst-30200	78	30	control	control	NOUN
ajst-30200	78	31	tasks	task	NOUN
ajst-30200	78	32	.	.	PUNCT
ajst-30200	79	1	2	2	X
ajst-30200	79	2	.	.	X
ajst-30200	79	3	policy	policy	NOUN
ajst-30200	79	4	gradient	gradient	NOUN
ajst-30200	79	5	methods	method	NOUN
ajst-30200	79	6	and	and	CCONJ
ajst-30200	79	7	actor	actor	NOUN
ajst-30200	79	8	-	-	PUNCT
ajst-30200	79	9	critic	critic	NOUN
ajst-30200	79	10	methods	method	NOUN
ajst-30200	79	11	policy	policy	NOUN
ajst-30200	79	12	gradient	gradient	NOUN
ajst-30200	79	13	methods	method	NOUN
ajst-30200	79	14	directly	directly	ADV
ajst-30200	79	15	optimize	optimize	VERB
ajst-30200	79	16	the	the	DET
ajst-30200	79	17	parameters	parameter	NOUN
ajst-30200	79	18	of	of	ADP
ajst-30200	79	19	the	the	DET
ajst-30200	79	20	policy	policy	NOUN
ajst-30200	79	21	.	.	PUNCT
ajst-30200	80	1	traditional	traditional	ADJ
ajst-30200	80	2	value	value	NOUN
ajst-30200	80	3	function	function	NOUN
ajst-30200	80	4	methods	method	NOUN
ajst-30200	80	5	infer	infer	VERB
ajst-30200	80	6	the	the	DET
ajst-30200	80	7	optimal	optimal	ADJ
ajst-30200	80	8	policy	policy	NOUN
ajst-30200	80	9	indirectly	indirectly	ADV
ajst-30200	80	10	by	by	ADP
ajst-30200	80	11	learning	learn	VERB
ajst-30200	80	12	q	q	NOUN
ajst-30200	80	13	-	-	NOUN
ajst-30200	80	14	values	value	NOUN
ajst-30200	80	15	,	,	PUNCT
ajst-30200	80	16	whereas	whereas	SCONJ
ajst-30200	80	17	policy	policy	NOUN
ajst-30200	80	18	gradient	gradient	NOUN
ajst-30200	80	19	methods	method	NOUN
ajst-30200	80	20	directly	directly	ADV
ajst-30200	80	21	optimize	optimize	VERB
ajst-30200	80	22	the	the	DET
ajst-30200	80	23	policy	policy	NOUN
ajst-30200	80	24	function	function	NOUN
ajst-30200	80	25	,	,	PUNCT
ajst-30200	80	26	compute	compute	NOUN
ajst-30200	80	27	gradients	gradient	NOUN
ajst-30200	80	28	,	,	PUNCT
ajst-30200	80	29	and	and	CCONJ
ajst-30200	80	30	update	update	NOUN
ajst-30200	80	31	parameters	parameter	NOUN
ajst-30200	80	32	.	.	PUNCT
ajst-30200	81	1	by	by	ADP
ajst-30200	81	2	backpropagating	backpropagate	VERB
ajst-30200	81	3	the	the	DET
ajst-30200	81	4	gradients	gradient	NOUN
ajst-30200	81	5	,	,	PUNCT
ajst-30200	81	6	the	the	DET
ajst-30200	81	7	agent	agent	NOUN
ajst-30200	81	8	can	can	AUX
ajst-30200	81	9	effectively	effectively	ADV
ajst-30200	81	10	improve	improve	VERB
ajst-30200	81	11	its	its	PRON
ajst-30200	81	12	behavior	behavior	NOUN
ajst-30200	81	13	strategy	strategy	NOUN
ajst-30200	81	14	.	.	PUNCT
ajst-30200	82	1	policy	policy	NOUN
ajst-30200	82	2	gradient	gradient	NOUN
ajst-30200	82	3	methods	method	NOUN
ajst-30200	82	4	are	be	AUX
ajst-30200	82	5	particularly	particularly	ADV
ajst-30200	82	6	effective	effective	ADJ
ajst-30200	82	7	for	for	ADP
ajst-30200	82	8	high	high	ADJ
ajst-30200	82	9	-	-	PUNCT
ajst-30200	82	10	dimensional	dimensional	ADJ
ajst-30200	82	11	continuous	continuous	ADJ
ajst-30200	82	12	action	action	NOUN
ajst-30200	82	13	spaces	space	NOUN
ajst-30200	82	14	.	.	PUNCT
ajst-30200	83	1	the	the	DET
ajst-30200	83	2	actor	actor	NOUN
ajst-30200	83	3	-	-	PUNCT
ajst-30200	83	4	critic	critic	NOUN
ajst-30200	83	5	method	method	NOUN
ajst-30200	83	6	combines	combine	VERB
ajst-30200	83	7	value	value	NOUN
ajst-30200	83	8	function	function	NOUN
ajst-30200	83	9	methods	method	NOUN
ajst-30200	83	10	and	and	CCONJ
ajst-30200	83	11	policy	policy	NOUN
ajst-30200	83	12	gradient	gradient	NOUN
ajst-30200	83	13	methods	method	NOUN
ajst-30200	83	14	,	,	PUNCT
ajst-30200	83	15	overcoming	overcome	VERB
ajst-30200	83	16	the	the	DET
ajst-30200	83	17	limitations	limitation	NOUN
ajst-30200	83	18	of	of	ADP
ajst-30200	83	19	both	both	PRON
ajst-30200	83	20	.	.	PUNCT
ajst-30200	84	1	in	in	ADP
ajst-30200	84	2	actor	actor	NOUN
ajst-30200	84	3	-	-	PUNCT
ajst-30200	84	4	critic	critic	NOUN
ajst-30200	84	5	,	,	PUNCT
ajst-30200	84	6	the	the	DET
ajst-30200	84	7	actor	actor	NOUN
ajst-30200	84	8	generates	generate	VERB
ajst-30200	84	9	the	the	DET
ajst-30200	84	10	action	action	NOUN
ajst-30200	84	11	policy	policy	NOUN
ajst-30200	84	12	,	,	PUNCT
ajst-30200	84	13	while	while	SCONJ
ajst-30200	84	14	the	the	DET
ajst-30200	84	15	critic	critic	NOUN
ajst-30200	84	16	is	be	AUX
ajst-30200	84	17	used	use	VERB
ajst-30200	84	18	to	to	PART
ajst-30200	84	19	evaluate	evaluate	VERB
ajst-30200	84	20	the	the	DET
ajst-30200	84	21	quality	quality	NOUN
ajst-30200	84	22	of	of	ADP
ajst-30200	84	23	the	the	DET
ajst-30200	84	24	current	current	ADJ
ajst-30200	84	25	policy	policy	NOUN
ajst-30200	84	26	and	and	CCONJ
ajst-30200	84	27	provides	provide	VERB
ajst-30200	84	28	feedback	feedback	NOUN
ajst-30200	84	29	to	to	ADP
ajst-30200	84	30	the	the	DET
ajst-30200	84	31	actor	actor	NOUN
ajst-30200	84	32	.	.	PUNCT
ajst-30200	85	1	the	the	DET
ajst-30200	85	2	actor	actor	NOUN
ajst-30200	85	3	adjusts	adjust	VERB
ajst-30200	85	4	its	its	PRON
ajst-30200	85	5	behavior	behavior	NOUN
ajst-30200	85	6	strategy	strategy	NOUN
ajst-30200	85	7	based	base	VERB
ajst-30200	85	8	on	on	ADP
ajst-30200	85	9	the	the	DET
ajst-30200	85	10	critic	critic	NOUN
ajst-30200	85	11	’s	’s	PART
ajst-30200	85	12	evaluations	evaluation	NOUN
ajst-30200	85	13	,	,	PUNCT
ajst-30200	85	14	thereby	thereby	ADV
ajst-30200	85	15	improving	improve	VERB
ajst-30200	85	16	overall	overall	ADJ
ajst-30200	85	17	performance	performance	NOUN
ajst-30200	85	18	.	.	PUNCT
ajst-30200	86	1	the	the	DET
ajst-30200	86	2	advantage	advantage	NOUN
ajst-30200	86	3	of	of	ADP
ajst-30200	86	4	actorcritic	actorcritic	ADJ
ajst-30200	86	5	methods	method	NOUN
ajst-30200	86	6	is	be	AUX
ajst-30200	86	7	that	that	SCONJ
ajst-30200	86	8	they	they	PRON
ajst-30200	86	9	combine	combine	VERB
ajst-30200	86	10	policy	policy	NOUN
ajst-30200	86	11	optimization	optimization	NOUN
ajst-30200	86	12	and	and	CCONJ
ajst-30200	86	13	value	value	NOUN
ajst-30200	86	14	evaluation	evaluation	NOUN
ajst-30200	86	15	,	,	PUNCT
ajst-30200	86	16	making	make	VERB
ajst-30200	86	17	them	they	PRON
ajst-30200	86	18	suitable	suitable	ADJ
ajst-30200	86	19	for	for	ADP
ajst-30200	86	20	complex	complex	ADJ
ajst-30200	86	21	and	and	CCONJ
ajst-30200	86	22	dynamic	dynamic	ADJ
ajst-30200	86	23	environments[4	environments[4	NOUN
ajst-30200	86	24	]	]	X
ajst-30200	86	25	.	.	PUNCT
ajst-30200	87	1	3	3	X
ajst-30200	87	2	.	.	X
ajst-30200	87	3	reward	reward	NOUN
ajst-30200	87	4	design	design	NOUN
ajst-30200	87	5	and	and	CCONJ
ajst-30200	87	6	exploration	exploration	NOUN
ajst-30200	87	7	mechanisms	mechanism	NOUN
ajst-30200	87	8	in	in	ADP
ajst-30200	87	9	deep	deep	ADJ
ajst-30200	87	10	reinforcement	reinforcement	NOUN
ajst-30200	87	11	learning	learn	VERB
ajst-30200	87	12	reward	reward	NOUN
ajst-30200	87	13	design	design	NOUN
ajst-30200	87	14	is	be	AUX
ajst-30200	87	15	a	a	DET
ajst-30200	87	16	critical	critical	ADJ
ajst-30200	87	17	part	part	NOUN
ajst-30200	87	18	of	of	ADP
ajst-30200	87	19	reinforcement	reinforcement	NOUN
ajst-30200	87	20	learning	learning	NOUN
ajst-30200	87	21	,	,	PUNCT
ajst-30200	87	22	directly	directly	ADV
ajst-30200	87	23	influencing	influence	VERB
ajst-30200	87	24	the	the	DET
ajst-30200	87	25	agent	agent	NOUN
ajst-30200	87	26	’s	’s	PART
ajst-30200	87	27	learning	learning	NOUN
ajst-30200	87	28	process	process	NOUN
ajst-30200	87	29	and	and	CCONJ
ajst-30200	87	30	final	final	ADJ
ajst-30200	87	31	performance	performance	NOUN
ajst-30200	87	32	.	.	PUNCT
ajst-30200	88	1	a	a	DET
ajst-30200	88	2	well	well	ADV
ajst-30200	88	3	-	-	PUNCT
ajst-30200	88	4	designed	design	VERB
ajst-30200	88	5	reward	reward	NOUN
ajst-30200	88	6	function	function	NOUN
ajst-30200	88	7	can	can	AUX
ajst-30200	88	8	effectively	effectively	ADV
ajst-30200	88	9	guide	guide	VERB
ajst-30200	88	10	the	the	DET
ajst-30200	88	11	agent	agent	NOUN
ajst-30200	88	12	to	to	PART
ajst-30200	88	13	learn	learn	VERB
ajst-30200	88	14	the	the	DET
ajst-30200	88	15	desired	desire	VERB
ajst-30200	88	16	behavior	behavior	NOUN
ajst-30200	88	17	patterns	pattern	NOUN
ajst-30200	88	18	.	.	PUNCT
ajst-30200	89	1	the	the	DET
ajst-30200	89	2	reward	reward	NOUN
ajst-30200	89	3	function	function	NOUN
ajst-30200	89	4	needs	need	VERB
ajst-30200	89	5	to	to	PART
ajst-30200	89	6	account	account	VERB
ajst-30200	89	7	for	for	ADP
ajst-30200	89	8	the	the	DET
ajst-30200	89	9	characteristics	characteristic	NOUN
ajst-30200	89	10	of	of	ADP
ajst-30200	89	11	the	the	DET
ajst-30200	89	12	task	task	NOUN
ajst-30200	89	13	and	and	CCONJ
ajst-30200	89	14	the	the	DET
ajst-30200	89	15	objectives	objective	NOUN
ajst-30200	89	16	,	,	PUNCT
ajst-30200	89	17	avoiding	avoid	VERB
ajst-30200	89	18	over	over	ADV
ajst-30200	89	19	-	-	PUNCT
ajst-30200	89	20	rewarding	reward	VERB
ajst-30200	89	21	or	or	CCONJ
ajst-30200	89	22	overpunishing	overpunishe	VERB
ajst-30200	89	23	certain	certain	ADJ
ajst-30200	89	24	behaviors	behavior	NOUN
ajst-30200	89	25	.	.	PUNCT
ajst-30200	90	1	common	common	ADJ
ajst-30200	90	2	reward	reward	NOUN
ajst-30200	90	3	design	design	NOUN
ajst-30200	90	4	methods	method	NOUN
ajst-30200	90	5	include	include	VERB
ajst-30200	90	6	sparse	sparse	ADJ
ajst-30200	90	7	rewards	reward	NOUN
ajst-30200	90	8	and	and	CCONJ
ajst-30200	90	9	dense	dense	ADJ
ajst-30200	90	10	rewards	reward	NOUN
ajst-30200	90	11	,	,	PUNCT
ajst-30200	90	12	where	where	SCONJ
ajst-30200	90	13	sparse	sparse	ADJ
ajst-30200	90	14	rewards	reward	NOUN
ajst-30200	90	15	typically	typically	ADV
ajst-30200	90	16	require	require	VERB
ajst-30200	90	17	the	the	DET
ajst-30200	90	18	agent	agent	NOUN
ajst-30200	90	19	to	to	PART
ajst-30200	90	20	try	try	VERB
ajst-30200	90	21	continuously	continuously	ADV
ajst-30200	90	22	over	over	ADP
ajst-30200	90	23	a	a	DET
ajst-30200	90	24	long	long	ADJ
ajst-30200	90	25	period	period	NOUN
ajst-30200	90	26	to	to	PART
ajst-30200	90	27	obtain	obtain	VERB
ajst-30200	90	28	rewards	reward	NOUN
ajst-30200	90	29	,	,	PUNCT
ajst-30200	90	30	while	while	SCONJ
ajst-30200	90	31	dense	dense	ADJ
ajst-30200	90	32	rewards	reward	NOUN
ajst-30200	90	33	provide	provide	VERB
ajst-30200	90	34	feedback	feedback	NOUN
ajst-30200	90	35	for	for	ADP
ajst-30200	90	36	each	each	DET
ajst-30200	90	37	action	action	NOUN
ajst-30200	90	38	taken	take	VERB
ajst-30200	90	39	.	.	PUNCT
ajst-30200	91	1	exploration	exploration	NOUN
ajst-30200	91	2	mechanisms	mechanism	NOUN
ajst-30200	91	3	are	be	AUX
ajst-30200	91	4	another	another	DET
ajst-30200	91	5	key	key	ADJ
ajst-30200	91	6	element	element	NOUN
ajst-30200	91	7	of	of	ADP
ajst-30200	91	8	reinforcement	reinforcement	NOUN
ajst-30200	91	9	learning	learning	NOUN
ajst-30200	91	10	.	.	PUNCT
ajst-30200	92	1	the	the	DET
ajst-30200	92	2	agent	agent	NOUN
ajst-30200	92	3	needs	need	VERB
ajst-30200	92	4	to	to	PART
ajst-30200	92	5	balance	balance	VERB
ajst-30200	92	6	exploration	exploration	NOUN
ajst-30200	92	7	(	(	PUNCT
ajst-30200	92	8	trying	try	VERB
ajst-30200	92	9	new	new	ADJ
ajst-30200	92	10	actions	action	NOUN
ajst-30200	92	11	)	)	PUNCT
ajst-30200	92	12	with	with	ADP
ajst-30200	92	13	exploitation	exploitation	NOUN
ajst-30200	92	14	(	(	PUNCT
ajst-30200	92	15	choosing	choose	VERB
ajst-30200	92	16	the	the	DET
ajst-30200	92	17	best	well	ADV
ajst-30200	92	18	-	-	PUNCT
ajst-30200	92	19	known	know	VERB
ajst-30200	92	20	action	action	NOUN
ajst-30200	92	21	)	)	PUNCT
ajst-30200	92	22	.	.	PUNCT
ajst-30200	93	1	common	common	ADJ
ajst-30200	93	2	exploration	exploration	NOUN
ajst-30200	93	3	mechanisms	mechanism	NOUN
ajst-30200	93	4	include	include	VERB
ajst-30200	93	5	the	the	DET
ajst-30200	93	6	ε	ε	PROPN
ajst-30200	93	7	-	-	PUNCT
ajst-30200	93	8	greedy	greedy	ADJ
ajst-30200	93	9	strategy	strategy	NOUN
ajst-30200	93	10	,	,	PUNCT
ajst-30200	93	11	where	where	SCONJ
ajst-30200	93	12	the	the	DET
ajst-30200	93	13	agent	agent	NOUN
ajst-30200	93	14	has	have	VERB
ajst-30200	93	15	a	a	DET
ajst-30200	93	16	certain	certain	ADJ
ajst-30200	93	17	probability	probability	NOUN
ajst-30200	93	18	of	of	ADP
ajst-30200	93	19	selecting	select	VERB
ajst-30200	93	20	a	a	DET
ajst-30200	93	21	random	random	ADJ
ajst-30200	93	22	action	action	NOUN
ajst-30200	93	23	to	to	PART
ajst-30200	93	24	explore	explore	VERB
ajst-30200	93	25	,	,	PUNCT
ajst-30200	93	26	and	and	CCONJ
ajst-30200	93	27	the	the	DET
ajst-30200	93	28	remaining	remain	VERB
ajst-30200	93	29	time	time	NOUN
ajst-30200	93	30	it	it	PRON
ajst-30200	93	31	chooses	choose	VERB
ajst-30200	93	32	the	the	DET
ajst-30200	93	33	current	current	ADJ
ajst-30200	93	34	optimal	optimal	ADJ
ajst-30200	93	35	action	action	NOUN
ajst-30200	93	36	.	.	PUNCT
ajst-30200	94	1	other	other	ADJ
ajst-30200	94	2	exploration	exploration	NOUN
ajst-30200	94	3	strategies	strategy	NOUN
ajst-30200	94	4	,	,	PUNCT
ajst-30200	94	5	such	such	ADJ
ajst-30200	94	6	as	as	ADP
ajst-30200	94	7	the	the	DET
ajst-30200	94	8	softmax	softmax	NOUN
ajst-30200	94	9	method	method	NOUN
ajst-30200	94	10	,	,	PUNCT
ajst-30200	94	11	are	be	AUX
ajst-30200	94	12	also	also	ADV
ajst-30200	94	13	widely	widely	ADV
ajst-30200	94	14	used	use	VERB
ajst-30200	94	15	in	in	ADP
ajst-30200	94	16	deep	deep	ADJ
ajst-30200	94	17	reinforcement	reinforcement	NOUN
ajst-30200	94	18	learning	learning	NOUN
ajst-30200	94	19	.	.	PUNCT
ajst-30200	95	1	(	(	PUNCT
ajst-30200	95	2	c	c	X
ajst-30200	95	3	)	)	PUNCT
ajst-30200	95	4	applications	application	NOUN
ajst-30200	95	5	of	of	ADP
ajst-30200	95	6	deep	deep	ADJ
ajst-30200	95	7	reinforcement	reinforcement	NOUN
ajst-30200	95	8	learning	learning	NOUN
ajst-30200	95	9	in	in	ADP
ajst-30200	95	10	control	control	NOUN
ajst-30200	95	11	1	1	NUM
ajst-30200	95	12	.	.	PUNCT
ajst-30200	96	1	application	application	NOUN
ajst-30200	96	2	of	of	ADP
ajst-30200	96	3	deep	deep	ADJ
ajst-30200	96	4	reinforcement	reinforcement	NOUN
ajst-30200	96	5	learning	learning	NOUN
ajst-30200	96	6	in	in	ADP
ajst-30200	96	7	robot	robot	NOUN
ajst-30200	96	8	control	control	PROPN
ajst-30200	96	9	deep	deep	ADJ
ajst-30200	96	10	reinforcement	reinforcement	NOUN
ajst-30200	96	11	learning	learning	NOUN
ajst-30200	96	12	has	have	AUX
ajst-30200	96	13	made	make	VERB
ajst-30200	96	14	significant	significant	ADJ
ajst-30200	96	15	progress	progress	NOUN
ajst-30200	96	16	in	in	ADP
ajst-30200	96	17	robot	robot	NOUN
ajst-30200	96	18	control	control	PROPN
ajst-30200	96	19	applications	application	NOUN
ajst-30200	96	20	.	.	PUNCT
ajst-30200	97	1	traditional	traditional	ADJ
ajst-30200	97	2	robot	robot	NOUN
ajst-30200	97	3	control	control	NOUN
ajst-30200	97	4	methods	method	NOUN
ajst-30200	97	5	typically	typically	ADV
ajst-30200	97	6	rely	rely	VERB
ajst-30200	97	7	on	on	ADP
ajst-30200	97	8	manual	manual	ADJ
ajst-30200	97	9	tuning	tuning	NOUN
ajst-30200	97	10	and	and	CCONJ
ajst-30200	97	11	predefined	predefine	VERB
ajst-30200	97	12	rules	rule	NOUN
ajst-30200	97	13	,	,	PUNCT
ajst-30200	97	14	whereas	whereas	SCONJ
ajst-30200	97	15	deep	deep	ADJ
ajst-30200	97	16	reinforcement	reinforcement	NOUN
ajst-30200	97	17	learning	learning	NOUN
ajst-30200	97	18	autonomously	autonomously	ADV
ajst-30200	97	19	learns	learn	VERB
ajst-30200	97	20	control	control	NOUN
ajst-30200	97	21	strategies	strategy	NOUN
ajst-30200	97	22	through	through	ADP
ajst-30200	97	23	interactions	interaction	NOUN
ajst-30200	97	24	with	with	ADP
ajst-30200	97	25	the	the	DET
ajst-30200	97	26	environment	environment	NOUN
ajst-30200	97	27	.	.	PUNCT
ajst-30200	98	1	in	in	ADP
ajst-30200	98	2	robot	robot	NOUN
ajst-30200	98	3	control	control	NOUN
ajst-30200	98	4	,	,	PUNCT
ajst-30200	98	5	deep	deep	ADJ
ajst-30200	98	6	reinforcement	reinforcement	NOUN
ajst-30200	98	7	learning	learning	NOUN
ajst-30200	98	8	is	be	AUX
ajst-30200	98	9	applied	apply	VERB
ajst-30200	98	10	to	to	ADP
ajst-30200	98	11	tasks	task	NOUN
ajst-30200	98	12	such	such	ADJ
ajst-30200	98	13	as	as	ADP
ajst-30200	98	14	autonomous	autonomous	ADJ
ajst-30200	98	15	navigation	navigation	NOUN
ajst-30200	98	16	,	,	PUNCT
ajst-30200	98	17	object	object	VERB
ajst-30200	98	18	manipulation	manipulation	NOUN
ajst-30200	98	19	,	,	PUNCT
ajst-30200	98	20	and	and	CCONJ
ajst-30200	98	21	path	path	NOUN
ajst-30200	98	22	planning	planning	NOUN
ajst-30200	98	23	.	.	PUNCT
ajst-30200	99	1	through	through	ADP
ajst-30200	99	2	deep	deep	ADJ
ajst-30200	99	3	reinforcement	reinforcement	NOUN
ajst-30200	99	4	learning	learning	NOUN
ajst-30200	99	5	,	,	PUNCT
ajst-30200	99	6	robots	robot	NOUN
ajst-30200	99	7	can	can	AUX
ajst-30200	99	8	gradually	gradually	ADV
ajst-30200	99	9	improve	improve	VERB
ajst-30200	99	10	their	their	PRON
ajst-30200	99	11	control	control	NOUN
ajst-30200	99	12	strategies	strategy	NOUN
ajst-30200	99	13	based	base	VERB
ajst-30200	99	14	on	on	ADP
ajst-30200	99	15	feedback	feedback	NOUN
ajst-30200	99	16	in	in	ADP
ajst-30200	99	17	complex	complex	ADJ
ajst-30200	99	18	environments	environment	NOUN
ajst-30200	99	19	and	and	CCONJ
ajst-30200	99	20	adapt	adapt	VERB
ajst-30200	99	21	to	to	PART
ajst-30200	99	22	dynamically	dynamically	ADV
ajst-30200	99	23	changing	change	VERB
ajst-30200	99	24	task	task	NOUN
ajst-30200	99	25	requirements	requirement	NOUN
ajst-30200	99	26	.	.	PUNCT
ajst-30200	100	1	2	2	X
ajst-30200	100	2	.	.	X
ajst-30200	100	3	application	application	NOUN
ajst-30200	100	4	of	of	ADP
ajst-30200	100	5	deep	deep	ADJ
ajst-30200	100	6	reinforcement	reinforcement	NOUN
ajst-30200	100	7	learning	learning	NOUN
ajst-30200	100	8	in	in	ADP
ajst-30200	100	9	smart	smart	ADJ
ajst-30200	100	10	grid	grid	NOUN
ajst-30200	100	11	management	management	NOUN
ajst-30200	100	12	deep	deep	ADJ
ajst-30200	100	13	reinforcement	reinforcement	NOUN
ajst-30200	100	14	learning	learning	NOUN
ajst-30200	100	15	is	be	AUX
ajst-30200	100	16	widely	widely	ADV
ajst-30200	100	17	used	use	VERB
ajst-30200	100	18	in	in	ADP
ajst-30200	100	19	the	the	DET
ajst-30200	100	20	management	management	NOUN
ajst-30200	100	21	of	of	ADP
ajst-30200	100	22	smart	smart	ADJ
ajst-30200	100	23	grids	grid	NOUN
ajst-30200	100	24	for	for	ADP
ajst-30200	100	25	load	load	NOUN
ajst-30200	100	26	forecasting	forecasting	NOUN
ajst-30200	100	27	,	,	PUNCT
ajst-30200	100	28	demand	demand	NOUN
ajst-30200	100	29	response	response	NOUN
ajst-30200	100	30	,	,	PUNCT
ajst-30200	100	31	and	and	CCONJ
ajst-30200	100	32	energy	energy	NOUN
ajst-30200	100	33	scheduling	scheduling	NOUN
ajst-30200	100	34	.	.	PUNCT
ajst-30200	101	1	smart	smart	ADJ
ajst-30200	101	2	grids	grid	NOUN
ajst-30200	101	3	feature	feature	VERB
ajst-30200	101	4	highly	highly	ADV
ajst-30200	101	5	complex	complex	ADJ
ajst-30200	101	6	systems	system	NOUN
ajst-30200	101	7	and	and	CCONJ
ajst-30200	101	8	uncertainty	uncertainty	NOUN
ajst-30200	101	9	,	,	PUNCT
ajst-30200	101	10	making	make	VERB
ajst-30200	101	11	traditional	traditional	ADJ
ajst-30200	101	12	control	control	NOUN
ajst-30200	101	13	methods	method	NOUN
ajst-30200	101	14	difficult	difficult	ADJ
ajst-30200	101	15	to	to	PART
ajst-30200	101	16	apply	apply	VERB
ajst-30200	101	17	.	.	PUNCT
ajst-30200	102	1	by	by	ADP
ajst-30200	102	2	using	use	VERB
ajst-30200	102	3	deep	deep	ADJ
ajst-30200	102	4	reinforcement	reinforcement	NOUN
ajst-30200	102	5	learning	learning	NOUN
ajst-30200	102	6	,	,	PUNCT
ajst-30200	102	7	the	the	DET
ajst-30200	102	8	smart	smart	ADJ
ajst-30200	102	9	grid	grid	NOUN
ajst-30200	102	10	can	can	AUX
ajst-30200	102	11	automatically	automatically	ADV
ajst-30200	102	12	adjust	adjust	VERB
ajst-30200	102	13	its	its	PRON
ajst-30200	102	14	operation	operation	NOUN
ajst-30200	102	15	strategy	strategy	NOUN
ajst-30200	102	16	based	base	VERB
ajst-30200	102	17	on	on	ADP
ajst-30200	102	18	real	real	ADJ
ajst-30200	102	19	-	-	PUNCT
ajst-30200	102	20	time	time	NOUN
ajst-30200	102	21	power	power	NOUN
ajst-30200	102	22	demand	demand	NOUN
ajst-30200	102	23	and	and	CCONJ
ajst-30200	102	24	supply	supply	NOUN
ajst-30200	102	25	conditions	condition	NOUN
ajst-30200	102	26	,	,	PUNCT
ajst-30200	102	27	thereby	thereby	ADV
ajst-30200	102	28	achieving	achieve	VERB
ajst-30200	102	29	efficient	efficient	ADJ
ajst-30200	102	30	energy	energy	NOUN
ajst-30200	102	31	distribution	distribution	NOUN
ajst-30200	102	32	and	and	CCONJ
ajst-30200	102	33	utilization	utilization	NOUN
ajst-30200	102	34	.	.	PUNCT
ajst-30200	103	1	deep	deep	ADJ
ajst-30200	103	2	reinforcement	reinforcement	NOUN
ajst-30200	103	3	learning	learning	NOUN
ajst-30200	103	4	can	can	AUX
ajst-30200	103	5	also	also	ADV
ajst-30200	103	6	optimize	optimize	VERB
ajst-30200	103	7	power	power	NOUN
ajst-30200	103	8	system	system	NOUN
ajst-30200	103	9	scheduling	scheduling	NOUN
ajst-30200	103	10	,	,	PUNCT
ajst-30200	103	11	reducing	reduce	VERB
ajst-30200	103	12	energy	energy	NOUN
ajst-30200	103	13	consumption	consumption	NOUN
ajst-30200	103	14	and	and	CCONJ
ajst-30200	103	15	operational	operational	ADJ
ajst-30200	103	16	costs	cost	NOUN
ajst-30200	103	17	.	.	PUNCT
ajst-30200	104	1	3	3	X
ajst-30200	104	2	.	.	X
ajst-30200	104	3	application	application	NOUN
ajst-30200	104	4	of	of	ADP
ajst-30200	104	5	deep	deep	ADJ
ajst-30200	104	6	reinforcement	reinforcement	NOUN
ajst-30200	104	7	learning	learning	NOUN
ajst-30200	104	8	in	in	ADP
ajst-30200	104	9	wave	wave	NOUN
ajst-30200	104	10	energy	energy	NOUN
ajst-30200	104	11	device	device	NOUN
ajst-30200	104	12	control	control	NOUN
ajst-30200	104	13	in	in	ADP
ajst-30200	104	14	the	the	DET
ajst-30200	104	15	control	control	NOUN
ajst-30200	104	16	of	of	ADP
ajst-30200	104	17	wave	wave	NOUN
ajst-30200	104	18	energy	energy	NOUN
ajst-30200	104	19	devices	device	NOUN
ajst-30200	104	20	,	,	PUNCT
ajst-30200	104	21	deep	deep	ADJ
ajst-30200	104	22	reinforcement	reinforcement	NOUN
ajst-30200	104	23	learning	learning	NOUN
ajst-30200	104	24	is	be	AUX
ajst-30200	104	25	used	use	VERB
ajst-30200	104	26	to	to	PART
ajst-30200	104	27	optimize	optimize	VERB
ajst-30200	104	28	the	the	DET
ajst-30200	104	29	energy	energy	NOUN
ajst-30200	104	30	capture	capture	NOUN
ajst-30200	104	31	process	process	NOUN
ajst-30200	104	32	and	and	CCONJ
ajst-30200	104	33	improve	improve	VERB
ajst-30200	104	34	the	the	DET
ajst-30200	104	35	stability	stability	NOUN
ajst-30200	104	36	of	of	ADP
ajst-30200	104	37	the	the	DET
ajst-30200	104	38	device	device	NOUN
ajst-30200	104	39	.	.	PUNCT
ajst-30200	105	1	the	the	DET
ajst-30200	105	2	energy	energy	NOUN
ajst-30200	105	3	conversion	conversion	NOUN
ajst-30200	105	4	efficiency	efficiency	NOUN
ajst-30200	105	5	of	of	ADP
ajst-30200	105	6	an	an	DET
ajst-30200	105	7	oscillating	oscillate	VERB
ajst-30200	105	8	water	water	NOUN
ajst-30200	105	9	column	column	NOUN
ajst-30200	105	10	wave	wave	NOUN
ajst-30200	105	11	energy	energy	NOUN
ajst-30200	105	12	device	device	NOUN
ajst-30200	105	13	is	be	AUX
ajst-30200	105	14	influenced	influence	VERB
ajst-30200	105	15	by	by	ADP
ajst-30200	105	16	factors	factor	NOUN
ajst-30200	105	17	such	such	ADJ
ajst-30200	105	18	as	as	ADP
ajst-30200	105	19	wave	wave	NOUN
ajst-30200	105	20	variation	variation	NOUN
ajst-30200	105	21	,	,	PUNCT
ajst-30200	105	22	system	system	NOUN
ajst-30200	105	23	dynamic	dynamic	ADJ
ajst-30200	105	24	response	response	NOUN
ajst-30200	105	25	,	,	PUNCT
ajst-30200	105	26	and	and	CCONJ
ajst-30200	105	27	turbine	turbine	NOUN
ajst-30200	105	28	control	control	NOUN
ajst-30200	105	29	strategy	strategy	NOUN
ajst-30200	105	30	.	.	PUNCT
ajst-30200	106	1	by	by	ADP
ajst-30200	106	2	introducing	introduce	VERB
ajst-30200	106	3	deep	deep	ADJ
ajst-30200	106	4	reinforcement	reinforcement	NOUN
ajst-30200	106	5	learning	learning	NOUN
ajst-30200	106	6	,	,	PUNCT
ajst-30200	106	7	the	the	DET
ajst-30200	106	8	agent	agent	NOUN
ajst-30200	106	9	can	can	AUX
ajst-30200	106	10	automatically	automatically	ADV
ajst-30200	106	11	learn	learn	VERB
ajst-30200	106	12	the	the	DET
ajst-30200	106	13	optimal	optimal	ADJ
ajst-30200	106	14	turbine	turbine	NOUN
ajst-30200	106	15	control	control	NOUN
ajst-30200	106	16	strategy	strategy	NOUN
ajst-30200	106	17	under	under	ADP
ajst-30200	106	18	dynamic	dynamic	ADJ
ajst-30200	106	19	wave	wave	NOUN
ajst-30200	106	20	conditions	condition	NOUN
ajst-30200	106	21	,	,	PUNCT
ajst-30200	106	22	adjusting	adjust	VERB
ajst-30200	106	23	the	the	DET
ajst-30200	106	24	device	device	NOUN
ajst-30200	106	25	's	's	PART
ajst-30200	106	26	working	work	VERB
ajst-30200	106	27	state	state	NOUN
ajst-30200	106	28	in	in	ADP
ajst-30200	106	29	real	real	ADJ
ajst-30200	106	30	-	-	PUNCT
ajst-30200	106	31	time	time	NOUN
ajst-30200	106	32	to	to	PART
ajst-30200	106	33	improve	improve	VERB
ajst-30200	106	34	energy	energy	NOUN
ajst-30200	106	35	conversion	conversion	NOUN
ajst-30200	106	36	efficiency	efficiency	NOUN
ajst-30200	106	37	and	and	CCONJ
ajst-30200	106	38	reduce	reduce	VERB
ajst-30200	106	39	energy	energy	NOUN
ajst-30200	106	40	loss	loss	NOUN
ajst-30200	106	41	.	.	PUNCT
ajst-30200	107	1	furthermore	furthermore	ADV
ajst-30200	107	2	,	,	PUNCT
ajst-30200	107	3	deep	deep	ADJ
ajst-30200	107	4	reinforcement	reinforcement	NOUN
ajst-30200	107	5	learning	learning	NOUN
ajst-30200	107	6	can	can	AUX
ajst-30200	107	7	continually	continually	ADV
ajst-30200	107	8	optimize	optimize	VERB
ajst-30200	107	9	the	the	DET
ajst-30200	107	10	control	control	NOUN
ajst-30200	107	11	strategy	strategy	NOUN
ajst-30200	107	12	through	through	ADP
ajst-30200	107	13	feedback	feedback	NOUN
ajst-30200	107	14	mechanisms	mechanism	NOUN
ajst-30200	107	15	during	during	ADP
ajst-30200	107	16	the	the	DET
ajst-30200	107	17	long	long	ADJ
ajst-30200	107	18	-	-	PUNCT
ajst-30200	107	19	term	term	NOUN
ajst-30200	107	20	operation	operation	NOUN
ajst-30200	107	21	of	of	ADP
ajst-30200	107	22	the	the	DET
ajst-30200	107	23	wave	wave	NOUN
ajst-30200	107	24	energy	energy	NOUN
ajst-30200	107	25	device	device	NOUN
ajst-30200	107	26	,	,	PUNCT
ajst-30200	107	27	further	far	ADV
ajst-30200	107	28	enhancing	enhance	VERB
ajst-30200	107	29	system	system	NOUN
ajst-30200	107	30	stability	stability	NOUN
ajst-30200	107	31	and	and	CCONJ
ajst-30200	107	32	adaptability	adaptability	NOUN
ajst-30200	107	33	.	.	PUNCT
ajst-30200	108	1	3	3	X
ajst-30200	108	2	.	.	X
ajst-30200	108	3	adaptive	adaptive	PROPN
ajst-30200	108	4	control	control	PROPN
ajst-30200	108	5	strategy	strategy	NOUN
ajst-30200	108	6	design	design	NOUN
ajst-30200	108	7	based	base	VERB
ajst-30200	108	8	on	on	ADP
ajst-30200	108	9	deep	deep	ADJ
ajst-30200	108	10	reinforcement	reinforcement	NOUN
ajst-30200	108	11	learning	learning	NOUN
ajst-30200	108	12	(	(	PUNCT
ajst-30200	108	13	a	a	PRON
ajst-30200	108	14	)	)	PUNCT
ajst-30200	108	15	requirements	requirement	NOUN
ajst-30200	108	16	and	and	CCONJ
ajst-30200	108	17	goals	goal	NOUN
ajst-30200	108	18	of	of	ADP
ajst-30200	108	19	adaptive	adaptive	ADJ
ajst-30200	108	20	control	control	NOUN
ajst-30200	108	21	strategies	strategy	NOUN
ajst-30200	108	22	1	1	NUM
ajst-30200	108	23	.	.	PUNCT
ajst-30200	108	24	basic	basic	ADJ
ajst-30200	108	25	principles	principle	NOUN
ajst-30200	108	26	of	of	ADP
ajst-30200	108	27	adaptive	adaptive	ADJ
ajst-30200	108	28	control	control	NOUN
ajst-30200	108	29	adaptive	adaptive	ADJ
ajst-30200	108	30	control	control	NOUN
ajst-30200	108	31	is	be	AUX
ajst-30200	108	32	a	a	DET
ajst-30200	108	33	control	control	NOUN
ajst-30200	108	34	strategy	strategy	NOUN
ajst-30200	108	35	that	that	PRON
ajst-30200	108	36	automatically	automatically	ADV
ajst-30200	108	37	adjusts	adjust	VERB
ajst-30200	108	38	control	control	NOUN
ajst-30200	108	39	parameters	parameter	NOUN
ajst-30200	108	40	based	base	VERB
ajst-30200	108	41	on	on	ADP
ajst-30200	108	42	the	the	DET
ajst-30200	108	43	dynamic	dynamic	ADJ
ajst-30200	108	44	characteristics	characteristic	NOUN
ajst-30200	108	45	of	of	ADP
ajst-30200	108	46	the	the	DET
ajst-30200	108	47	system	system	NOUN
ajst-30200	108	48	.	.	PUNCT
ajst-30200	109	1	unlike	unlike	ADP
ajst-30200	109	2	traditional	traditional	ADJ
ajst-30200	109	3	fixedparameter	fixedparameter	ADJ
ajst-30200	109	4	control	control	PROPN
ajst-30200	109	5	methods	method	NOUN
ajst-30200	109	6	,	,	PUNCT
ajst-30200	109	7	adaptive	adaptive	ADJ
ajst-30200	109	8	control	control	NOUN
ajst-30200	109	9	can	can	AUX
ajst-30200	109	10	dynamically	dynamically	ADV
ajst-30200	109	11	adjust	adjust	VERB
ajst-30200	109	12	its	its	PRON
ajst-30200	109	13	behavior	behavior	NOUN
ajst-30200	109	14	strategy	strategy	NOUN
ajst-30200	109	15	in	in	ADP
ajst-30200	109	16	real	real	ADJ
ajst-30200	109	17	-	-	PUNCT
ajst-30200	109	18	time	time	NOUN
ajst-30200	109	19	according	accord	VERB
ajst-30200	109	20	to	to	ADP
ajst-30200	109	21	changes	change	NOUN
ajst-30200	109	22	in	in	ADP
ajst-30200	109	23	system	system	NOUN
ajst-30200	109	24	output	output	NOUN
ajst-30200	109	25	and	and	CCONJ
ajst-30200	109	26	environmental	environmental	ADJ
ajst-30200	109	27	conditions	condition	NOUN
ajst-30200	109	28	to	to	PART
ajst-30200	109	29	maintain	maintain	VERB
ajst-30200	109	30	optimal	optimal	ADJ
ajst-30200	109	31	system	system	NOUN
ajst-30200	109	32	performance	performance	NOUN
ajst-30200	109	33	under	under	ADP
ajst-30200	109	34	uncertainty	uncertainty	NOUN
ajst-30200	109	35	and	and	CCONJ
ajst-30200	109	36	varying	vary	VERB
ajst-30200	109	37	conditions	condition	NOUN
ajst-30200	109	38	.	.	PUNCT
ajst-30200	110	1	in	in	ADP
ajst-30200	110	2	control	control	PROPN
ajst-30200	110	3	theory	theory	NOUN
ajst-30200	110	4	,	,	PUNCT
ajst-30200	110	5	adaptive	adaptive	ADJ
ajst-30200	110	6	control	control	NOUN
ajst-30200	110	7	focuses	focus	VERB
ajst-30200	110	8	not	not	PART
ajst-30200	110	9	only	only	ADV
ajst-30200	110	10	on	on	ADP
ajst-30200	110	11	achieving	achieve	VERB
ajst-30200	110	12	system	system	NOUN
ajst-30200	110	13	stability	stability	NOUN
ajst-30200	110	14	but	but	CCONJ
ajst-30200	110	15	also	also	ADV
ajst-30200	110	16	on	on	ADP
ajst-30200	110	17	continuously	continuously	ADV
ajst-30200	110	18	optimizing	optimize	VERB
ajst-30200	110	19	control	control	NOUN
ajst-30200	110	20	parameters	parameter	NOUN
ajst-30200	110	21	through	through	ADP
ajst-30200	110	22	online	online	ADJ
ajst-30200	110	23	learning	learning	NOUN
ajst-30200	110	24	to	to	PART
ajst-30200	110	25	adapt	adapt	VERB
ajst-30200	110	26	to	to	ADP
ajst-30200	110	27	unknown	unknown	ADJ
ajst-30200	110	28	or	or	CCONJ
ajst-30200	110	29	changing	change	VERB
ajst-30200	110	30	environments	environment	NOUN
ajst-30200	110	31	.	.	PUNCT
ajst-30200	111	1	an	an	DET
ajst-30200	111	2	adaptive	adaptive	ADJ
ajst-30200	111	3	control	control	NOUN
ajst-30200	111	4	system	system	NOUN
ajst-30200	111	5	typically	typically	ADV
ajst-30200	111	6	includes	include	VERB
ajst-30200	111	7	system	system	NOUN
ajst-30200	111	8	identification	identification	NOUN
ajst-30200	111	9	,	,	PUNCT
ajst-30200	111	10	control	control	NOUN
ajst-30200	111	11	law	law	NOUN
ajst-30200	111	12	adjustment	adjustment	NOUN
ajst-30200	111	13	,	,	PUNCT
ajst-30200	111	14	and	and	CCONJ
ajst-30200	111	15	feedback	feedback	NOUN
ajst-30200	111	16	mechanisms	mechanism	NOUN
ajst-30200	111	17	.	.	PUNCT
ajst-30200	112	1	the	the	DET
ajst-30200	112	2	system	system	NOUN
ajst-30200	112	3	identification	identification	NOUN
ajst-30200	112	4	phase	phase	NOUN
ajst-30200	112	5	involves	involve	VERB
ajst-30200	112	6	realtime	realtime	ADJ
ajst-30200	112	7	collection	collection	NOUN
ajst-30200	112	8	of	of	ADP
ajst-30200	112	9	system	system	NOUN
ajst-30200	112	10	input	input	NOUN
ajst-30200	112	11	and	and	CCONJ
ajst-30200	112	12	output	output	NOUN
ajst-30200	112	13	data	datum	NOUN
ajst-30200	112	14	to	to	PART
ajst-30200	112	15	estimate	estimate	VERB
ajst-30200	112	16	the	the	DET
ajst-30200	112	17	system	system	NOUN
ajst-30200	112	18	's	's	PART
ajst-30200	112	19	parameters	parameter	NOUN
ajst-30200	112	20	or	or	CCONJ
ajst-30200	112	21	model	model	NOUN
ajst-30200	112	22	;	;	PUNCT
ajst-30200	112	23	the	the	DET
ajst-30200	112	24	control	control	NOUN
ajst-30200	112	25	law	law	NOUN
ajst-30200	112	26	adjustment	adjustment	NOUN
ajst-30200	112	27	phase	phase	NOUN
ajst-30200	112	28	then	then	ADV
ajst-30200	112	29	adjusts	adjust	VERB
ajst-30200	112	30	the	the	DET
ajst-30200	112	31	control	control	NOUN
ajst-30200	112	32	strategy	strategy	NOUN
ajst-30200	112	33	based	base	VERB
ajst-30200	112	34	on	on	ADP
ajst-30200	112	35	the	the	DET
ajst-30200	112	36	information	information	NOUN
ajst-30200	112	37	derived	derive	VERB
ajst-30200	112	38	from	from	ADP
ajst-30200	112	39	identification	identification	NOUN
ajst-30200	112	40	,	,	PUNCT
ajst-30200	112	41	making	make	VERB
ajst-30200	112	42	system	system	NOUN
ajst-30200	112	43	behavior	behavior	NOUN
ajst-30200	112	44	align	align	VERB
ajst-30200	112	45	with	with	ADP
ajst-30200	112	46	the	the	DET
ajst-30200	112	47	desired	desire	VERB
ajst-30200	112	48	goals	goal	NOUN
ajst-30200	112	49	.	.	PUNCT
ajst-30200	113	1	in	in	ADP
ajst-30200	113	2	the	the	DET
ajst-30200	113	3	context	context	NOUN
ajst-30200	113	4	of	of	ADP
ajst-30200	113	5	deep	deep	ADJ
ajst-30200	113	6	reinforcement	reinforcement	NOUN
ajst-30200	113	7	learning	learning	NOUN
ajst-30200	113	8	,	,	PUNCT
ajst-30200	113	9	adaptive	adaptive	ADJ
ajst-30200	113	10	control	control	NOUN
ajst-30200	113	11	models	model	VERB
ajst-30200	113	12	the	the	DET
ajst-30200	113	13	system	system	NOUN
ajst-30200	113	14	using	use	VERB
ajst-30200	113	15	deep	deep	ADJ
ajst-30200	113	16	learning	learning	NOUN
ajst-30200	113	17	networks	network	NOUN
ajst-30200	113	18	and	and	CCONJ
ajst-30200	113	19	optimizes	optimize	VERB
ajst-30200	113	20	the	the	DET
ajst-30200	113	21	strategy	strategy	NOUN
ajst-30200	113	22	using	use	VERB
ajst-30200	113	23	reinforcement	reinforcement	NOUN
ajst-30200	113	24	learning	learning	NOUN
ajst-30200	113	25	algorithms	algorithm	NOUN
ajst-30200	113	26	,	,	PUNCT
ajst-30200	113	27	allowing	allow	VERB
ajst-30200	113	28	the	the	DET
ajst-30200	113	29	control	control	NOUN
ajst-30200	113	30	201	201	NUM
ajst-30200	113	31	system	system	NOUN
ajst-30200	113	32	to	to	PART
ajst-30200	113	33	adjust	adjust	VERB
ajst-30200	113	34	and	and	CCONJ
ajst-30200	113	35	optimize	optimize	VERB
ajst-30200	113	36	effectively	effectively	ADV
ajst-30200	113	37	under	under	ADP
ajst-30200	113	38	different	different	ADJ
ajst-30200	113	39	operating	operating	NOUN
ajst-30200	113	40	conditions	condition	NOUN
ajst-30200	113	41	.	.	PUNCT
ajst-30200	114	1	2	2	X
ajst-30200	114	2	.	.	X
ajst-30200	114	3	control	control	NOUN
ajst-30200	114	4	requirements	requirement	NOUN
ajst-30200	114	5	of	of	ADP
ajst-30200	114	6	oscillating	oscillate	VERB
ajst-30200	114	7	water	water	NOUN
ajst-30200	114	8	column	column	NOUN
ajst-30200	114	9	wave	wave	NOUN
ajst-30200	114	10	energy	energy	NOUN
ajst-30200	114	11	devices	device	NOUN
ajst-30200	114	12	an	an	DET
ajst-30200	114	13	oscillating	oscillate	VERB
ajst-30200	114	14	water	water	NOUN
ajst-30200	114	15	column	column	NOUN
ajst-30200	114	16	wave	wave	NOUN
ajst-30200	114	17	energy	energy	NOUN
ajst-30200	114	18	device	device	NOUN
ajst-30200	114	19	is	be	AUX
ajst-30200	114	20	a	a	DET
ajst-30200	114	21	highly	highly	ADV
ajst-30200	114	22	wave	wave	NOUN
ajst-30200	114	23	-	-	PUNCT
ajst-30200	114	24	dependent	dependent	ADJ
ajst-30200	114	25	energy	energy	NOUN
ajst-30200	114	26	conversion	conversion	NOUN
ajst-30200	114	27	system	system	NOUN
ajst-30200	114	28	.	.	PUNCT
ajst-30200	115	1	the	the	DET
ajst-30200	115	2	control	control	NOUN
ajst-30200	115	3	objectives	objective	NOUN
ajst-30200	115	4	include	include	VERB
ajst-30200	115	5	optimizing	optimize	VERB
ajst-30200	115	6	energy	energy	NOUN
ajst-30200	115	7	conversion	conversion	NOUN
ajst-30200	115	8	efficiency	efficiency	NOUN
ajst-30200	115	9	under	under	ADP
ajst-30200	115	10	varying	vary	VERB
ajst-30200	115	11	wave	wave	NOUN
ajst-30200	115	12	conditions	condition	NOUN
ajst-30200	115	13	,	,	PUNCT
ajst-30200	115	14	maintaining	maintain	VERB
ajst-30200	115	15	long	long	ADJ
ajst-30200	115	16	-	-	PUNCT
ajst-30200	115	17	term	term	NOUN
ajst-30200	115	18	stable	stable	ADJ
ajst-30200	115	19	operation	operation	NOUN
ajst-30200	115	20	of	of	ADP
ajst-30200	115	21	the	the	DET
ajst-30200	115	22	device	device	NOUN
ajst-30200	115	23	,	,	PUNCT
ajst-30200	115	24	and	and	CCONJ
ajst-30200	115	25	preventing	prevent	VERB
ajst-30200	115	26	damage	damage	NOUN
ajst-30200	115	27	under	under	ADP
ajst-30200	115	28	extreme	extreme	ADJ
ajst-30200	115	29	sea	sea	NOUN
ajst-30200	115	30	conditions	condition	NOUN
ajst-30200	115	31	.	.	PUNCT
ajst-30200	116	1	the	the	DET
ajst-30200	116	2	specific	specific	ADJ
ajst-30200	116	3	control	control	NOUN
ajst-30200	116	4	requirements	requirement	NOUN
ajst-30200	116	5	are	be	AUX
ajst-30200	116	6	as	as	SCONJ
ajst-30200	116	7	follows	follow	VERB
ajst-30200	116	8	:	:	PUNCT
ajst-30200	116	9	real	real	ADJ
ajst-30200	116	10	-	-	PUNCT
ajst-30200	116	11	time	time	NOUN
ajst-30200	116	12	adaptation	adaptation	NOUN
ajst-30200	116	13	to	to	PART
ajst-30200	116	14	wave	wave	VERB
ajst-30200	116	15	variations	variation	NOUN
ajst-30200	116	16	:	:	PUNCT
ajst-30200	116	17	wave	wave	NOUN
ajst-30200	116	18	height	height	NOUN
ajst-30200	116	19	,	,	PUNCT
ajst-30200	116	20	frequency	frequency	NOUN
ajst-30200	116	21	,	,	PUNCT
ajst-30200	116	22	and	and	CCONJ
ajst-30200	116	23	direction	direction	NOUN
ajst-30200	116	24	constantly	constantly	ADV
ajst-30200	116	25	change	change	VERB
ajst-30200	116	26	,	,	PUNCT
ajst-30200	116	27	and	and	CCONJ
ajst-30200	116	28	traditional	traditional	ADJ
ajst-30200	116	29	control	control	NOUN
ajst-30200	116	30	strategies	strategy	NOUN
ajst-30200	116	31	struggle	struggle	VERB
ajst-30200	116	32	to	to	PART
ajst-30200	116	33	handle	handle	VERB
ajst-30200	116	34	these	these	DET
ajst-30200	116	35	complex	complex	ADJ
ajst-30200	116	36	dynamic	dynamic	ADJ
ajst-30200	116	37	changes	change	NOUN
ajst-30200	116	38	.	.	PUNCT
ajst-30200	117	1	deep	deep	ADJ
ajst-30200	117	2	reinforcement	reinforcement	NOUN
ajst-30200	117	3	learning	learning	NOUN
ajst-30200	117	4	can	can	AUX
ajst-30200	117	5	adjust	adjust	VERB
ajst-30200	117	6	turbine	turbine	NOUN
ajst-30200	117	7	working	work	VERB
ajst-30200	117	8	parameters	parameter	NOUN
ajst-30200	117	9	in	in	ADP
ajst-30200	117	10	real	real	ADJ
ajst-30200	117	11	-	-	PUNCT
ajst-30200	117	12	time	time	NOUN
ajst-30200	117	13	according	accord	VERB
ajst-30200	117	14	to	to	ADP
ajst-30200	117	15	the	the	DET
ajst-30200	117	16	current	current	ADJ
ajst-30200	117	17	wave	wave	NOUN
ajst-30200	117	18	conditions	condition	NOUN
ajst-30200	117	19	,	,	PUNCT
ajst-30200	117	20	making	make	VERB
ajst-30200	117	21	the	the	DET
ajst-30200	117	22	system	system	NOUN
ajst-30200	117	23	adaptable	adaptable	ADJ
ajst-30200	117	24	to	to	PART
ajst-30200	117	25	wave	wave	VERB
ajst-30200	117	26	fluctuations	fluctuation	NOUN
ajst-30200	117	27	.	.	PUNCT
ajst-30200	118	1	optimization	optimization	NOUN
ajst-30200	118	2	of	of	ADP
ajst-30200	118	3	energy	energy	NOUN
ajst-30200	118	4	conversion	conversion	NOUN
ajst-30200	118	5	efficiency	efficiency	NOUN
ajst-30200	118	6	:	:	PUNCT
ajst-30200	118	7	the	the	DET
ajst-30200	118	8	efficiency	efficiency	NOUN
ajst-30200	118	9	of	of	ADP
ajst-30200	118	10	wave	wave	NOUN
ajst-30200	118	11	energy	energy	NOUN
ajst-30200	118	12	conversion	conversion	NOUN
ajst-30200	118	13	is	be	AUX
ajst-30200	118	14	influenced	influence	VERB
ajst-30200	118	15	by	by	ADP
ajst-30200	118	16	wave	wave	NOUN
ajst-30200	118	17	frequency	frequency	NOUN
ajst-30200	118	18	and	and	CCONJ
ajst-30200	118	19	amplitude	amplitude	NOUN
ajst-30200	118	20	.	.	PUNCT
ajst-30200	119	1	the	the	DET
ajst-30200	119	2	control	control	NOUN
ajst-30200	119	3	strategy	strategy	NOUN
ajst-30200	119	4	must	must	AUX
ajst-30200	119	5	adjust	adjust	VERB
ajst-30200	119	6	the	the	DET
ajst-30200	119	7	turbine	turbine	NOUN
ajst-30200	119	8	speed	speed	NOUN
ajst-30200	119	9	and	and	CCONJ
ajst-30200	119	10	other	other	ADJ
ajst-30200	119	11	parameters	parameter	NOUN
ajst-30200	119	12	under	under	ADP
ajst-30200	119	13	different	different	ADJ
ajst-30200	119	14	wave	wave	NOUN
ajst-30200	119	15	conditions	condition	NOUN
ajst-30200	119	16	to	to	PART
ajst-30200	119	17	maximize	maximize	VERB
ajst-30200	119	18	energy	energy	NOUN
ajst-30200	119	19	capture	capture	NOUN
ajst-30200	119	20	efficiency	efficiency	NOUN
ajst-30200	119	21	.	.	PUNCT
ajst-30200	120	1	ensuring	ensure	VERB
ajst-30200	120	2	system	system	NOUN
ajst-30200	120	3	stability	stability	NOUN
ajst-30200	120	4	and	and	CCONJ
ajst-30200	120	5	safety	safety	NOUN
ajst-30200	120	6	:	:	PUNCT
ajst-30200	120	7	wave	wave	NOUN
ajst-30200	120	8	energy	energy	NOUN
ajst-30200	120	9	devices	device	NOUN
ajst-30200	120	10	must	must	AUX
ajst-30200	120	11	withstand	withstand	VERB
ajst-30200	120	12	extreme	extreme	ADJ
ajst-30200	120	13	wave	wave	NOUN
ajst-30200	120	14	conditions	condition	NOUN
ajst-30200	120	15	,	,	PUNCT
ajst-30200	120	16	such	such	ADJ
ajst-30200	120	17	as	as	ADP
ajst-30200	120	18	storms	storm	NOUN
ajst-30200	120	19	or	or	CCONJ
ajst-30200	120	20	large	large	ADJ
ajst-30200	120	21	fluctuations	fluctuation	NOUN
ajst-30200	120	22	in	in	ADP
ajst-30200	120	23	sea	sea	NOUN
ajst-30200	120	24	waves	wave	NOUN
ajst-30200	120	25	.	.	PUNCT
ajst-30200	121	1	during	during	ADP
ajst-30200	121	2	these	these	DET
ajst-30200	121	3	conditions	condition	NOUN
ajst-30200	121	4	,	,	PUNCT
ajst-30200	121	5	precise	precise	ADJ
ajst-30200	121	6	control	control	NOUN
ajst-30200	121	7	strategies	strategy	NOUN
ajst-30200	121	8	are	be	AUX
ajst-30200	121	9	necessary	necessary	ADJ
ajst-30200	121	10	to	to	PART
ajst-30200	121	11	maintain	maintain	VERB
ajst-30200	121	12	device	device	NOUN
ajst-30200	121	13	stability	stability	NOUN
ajst-30200	121	14	and	and	CCONJ
ajst-30200	121	15	avoid	avoid	VERB
ajst-30200	121	16	excessive	excessive	ADJ
ajst-30200	121	17	oscillations	oscillation	NOUN
ajst-30200	121	18	or	or	CCONJ
ajst-30200	121	19	damage	damage	NOUN
ajst-30200	121	20	.	.	PUNCT
ajst-30200	122	1	3	3	X
ajst-30200	122	2	.	.	X
ajst-30200	122	3	advantages	advantage	NOUN
ajst-30200	122	4	of	of	ADP
ajst-30200	122	5	deep	deep	ADJ
ajst-30200	122	6	reinforcement	reinforcement	NOUN
ajst-30200	122	7	learning	learning	NOUN
ajst-30200	122	8	in	in	ADP
ajst-30200	122	9	adaptive	adaptive	ADJ
ajst-30200	122	10	control	control	NOUN
ajst-30200	122	11	deep	deep	ADJ
ajst-30200	122	12	reinforcement	reinforcement	NOUN
ajst-30200	122	13	learning	learning	NOUN
ajst-30200	122	14	(	(	PUNCT
ajst-30200	122	15	drl	drl	PROPN
ajst-30200	122	16	)	)	PUNCT
ajst-30200	122	17	offers	offer	VERB
ajst-30200	122	18	significant	significant	ADJ
ajst-30200	122	19	advantages	advantage	NOUN
ajst-30200	122	20	as	as	ADP
ajst-30200	122	21	an	an	DET
ajst-30200	122	22	adaptive	adaptive	ADJ
ajst-30200	122	23	control	control	NOUN
ajst-30200	122	24	method	method	NOUN
ajst-30200	122	25	.	.	PUNCT
ajst-30200	123	1	firstly	firstly	ADV
ajst-30200	123	2	,	,	PUNCT
ajst-30200	123	3	drl	drl	PROPN
ajst-30200	123	4	can	can	AUX
ajst-30200	123	5	automatically	automatically	ADV
ajst-30200	123	6	extract	extract	VERB
ajst-30200	123	7	key	key	ADJ
ajst-30200	123	8	features	feature	NOUN
ajst-30200	123	9	from	from	ADP
ajst-30200	123	10	raw	raw	ADJ
ajst-30200	123	11	state	state	NOUN
ajst-30200	123	12	data	datum	NOUN
ajst-30200	123	13	without	without	ADP
ajst-30200	123	14	relying	rely	VERB
ajst-30200	123	15	on	on	ADP
ajst-30200	123	16	precise	precise	ADJ
ajst-30200	123	17	physical	physical	ADJ
ajst-30200	123	18	models	model	NOUN
ajst-30200	123	19	.	.	PUNCT
ajst-30200	124	1	secondly	secondly	ADV
ajst-30200	124	2	,	,	PUNCT
ajst-30200	124	3	reinforcement	reinforcement	NOUN
ajst-30200	124	4	learning	learning	NOUN
ajst-30200	124	5	adjusts	adjust	VERB
ajst-30200	124	6	control	control	NOUN
ajst-30200	124	7	strategies	strategy	NOUN
ajst-30200	124	8	through	through	ADP
ajst-30200	124	9	continuous	continuous	ADJ
ajst-30200	124	10	interaction	interaction	NOUN
ajst-30200	124	11	with	with	ADP
ajst-30200	124	12	the	the	DET
ajst-30200	124	13	environment	environment	NOUN
ajst-30200	124	14	,	,	PUNCT
ajst-30200	124	15	enabling	enable	VERB
ajst-30200	124	16	the	the	DET
ajst-30200	124	17	system	system	NOUN
ajst-30200	124	18	to	to	PART
ajst-30200	124	19	respond	respond	VERB
ajst-30200	124	20	flexibly	flexibly	ADV
ajst-30200	124	21	to	to	ADP
ajst-30200	124	22	environmental	environmental	ADJ
ajst-30200	124	23	changes	change	NOUN
ajst-30200	124	24	and	and	CCONJ
ajst-30200	124	25	optimize	optimize	VERB
ajst-30200	124	26	control	control	NOUN
ajst-30200	124	27	decisions	decision	NOUN
ajst-30200	124	28	via	via	ADP
ajst-30200	124	29	reward	reward	NOUN
ajst-30200	124	30	and	and	CCONJ
ajst-30200	124	31	punishment	punishment	NOUN
ajst-30200	124	32	mechanisms	mechanism	NOUN
ajst-30200	124	33	.	.	PUNCT
ajst-30200	125	1	in	in	ADP
ajst-30200	125	2	wave	wave	NOUN
ajst-30200	125	3	energy	energy	NOUN
ajst-30200	125	4	device	device	NOUN
ajst-30200	125	5	control	control	NOUN
ajst-30200	125	6	,	,	PUNCT
ajst-30200	125	7	deep	deep	ADJ
ajst-30200	125	8	reinforcement	reinforcement	NOUN
ajst-30200	125	9	learning	learning	NOUN
ajst-30200	125	10	can	can	AUX
ajst-30200	125	11	allow	allow	VERB
ajst-30200	125	12	the	the	DET
ajst-30200	125	13	agent	agent	NOUN
ajst-30200	125	14	to	to	PART
ajst-30200	125	15	learn	learn	VERB
ajst-30200	125	16	the	the	DET
ajst-30200	125	17	optimal	optimal	ADJ
ajst-30200	125	18	control	control	NOUN
ajst-30200	125	19	strategy	strategy	NOUN
ajst-30200	125	20	for	for	ADP
ajst-30200	125	21	different	different	ADJ
ajst-30200	125	22	sea	sea	NOUN
ajst-30200	125	23	conditions	condition	NOUN
ajst-30200	125	24	through	through	ADP
ajst-30200	125	25	interaction	interaction	NOUN
ajst-30200	125	26	with	with	ADP
ajst-30200	125	27	the	the	DET
ajst-30200	125	28	environment	environment	NOUN
ajst-30200	125	29	.	.	PUNCT
ajst-30200	126	1	furthermore	furthermore	ADV
ajst-30200	126	2	,	,	PUNCT
ajst-30200	126	3	drl	drl	PROPN
ajst-30200	126	4	can	can	AUX
ajst-30200	126	5	continually	continually	ADV
ajst-30200	126	6	improve	improve	VERB
ajst-30200	126	7	strategies	strategy	NOUN
ajst-30200	126	8	through	through	ADP
ajst-30200	126	9	multiple	multiple	ADJ
ajst-30200	126	10	rounds	round	NOUN
ajst-30200	126	11	of	of	ADP
ajst-30200	126	12	training	training	NOUN
ajst-30200	126	13	,	,	PUNCT
ajst-30200	126	14	ensuring	ensure	VERB
ajst-30200	126	15	that	that	SCONJ
ajst-30200	126	16	wave	wave	NOUN
ajst-30200	126	17	energy	energy	NOUN
ajst-30200	126	18	devices	device	NOUN
ajst-30200	126	19	maintain	maintain	VERB
ajst-30200	126	20	high	high	ADJ
ajst-30200	126	21	energy	energy	NOUN
ajst-30200	126	22	conversion	conversion	NOUN
ajst-30200	126	23	efficiency	efficiency	NOUN
ajst-30200	126	24	under	under	ADP
ajst-30200	126	25	varying	vary	VERB
ajst-30200	126	26	wave	wave	NOUN
ajst-30200	126	27	conditions	condition	NOUN
ajst-30200	126	28	and	and	CCONJ
ajst-30200	126	29	stable	stable	ADJ
ajst-30200	126	30	operation	operation	NOUN
ajst-30200	126	31	under	under	ADP
ajst-30200	126	32	extreme	extreme	ADJ
ajst-30200	126	33	waves	wave	NOUN
ajst-30200	126	34	.	.	PUNCT
ajst-30200	127	1	(	(	PUNCT
ajst-30200	127	2	b	b	X
ajst-30200	127	3	)	)	PUNCT
ajst-30200	127	4	control	control	NOUN
ajst-30200	127	5	strategy	strategy	NOUN
ajst-30200	127	6	design	design	NOUN
ajst-30200	127	7	based	base	VERB
ajst-30200	127	8	on	on	ADP
ajst-30200	127	9	deep	deep	ADJ
ajst-30200	127	10	reinforcement	reinforcement	NOUN
ajst-30200	127	11	learning	learning	NOUN
ajst-30200	127	12	1	1	NUM
ajst-30200	127	13	.	.	PUNCT
ajst-30200	127	14	environmental	environmental	ADJ
ajst-30200	127	15	state	state	NOUN
ajst-30200	127	16	modeling	modeling	NOUN
ajst-30200	127	17	and	and	CCONJ
ajst-30200	127	18	feature	feature	NOUN
ajst-30200	127	19	extraction	extraction	NOUN
ajst-30200	127	20	the	the	DET
ajst-30200	127	21	design	design	NOUN
ajst-30200	127	22	of	of	ADP
ajst-30200	127	23	control	control	NOUN
ajst-30200	127	24	strategies	strategy	NOUN
ajst-30200	127	25	based	base	VERB
ajst-30200	127	26	on	on	ADP
ajst-30200	127	27	deep	deep	ADJ
ajst-30200	127	28	reinforcement	reinforcement	NOUN
ajst-30200	127	29	learning	learning	NOUN
ajst-30200	127	30	starts	start	VERB
ajst-30200	127	31	with	with	ADP
ajst-30200	127	32	modeling	model	VERB
ajst-30200	127	33	the	the	DET
ajst-30200	127	34	environment	environment	NOUN
ajst-30200	127	35	.	.	PUNCT
ajst-30200	128	1	in	in	ADP
ajst-30200	128	2	the	the	DET
ajst-30200	128	3	oscillating	oscillate	VERB
ajst-30200	128	4	water	water	NOUN
ajst-30200	128	5	column	column	NOUN
ajst-30200	128	6	wave	wave	VERB
ajst-30200	128	7	energy	energy	NOUN
ajst-30200	128	8	device	device	NOUN
ajst-30200	128	9	,	,	PUNCT
ajst-30200	128	10	the	the	DET
ajst-30200	128	11	environment	environment	NOUN
ajst-30200	128	12	state	state	NOUN
ajst-30200	128	13	includes	include	VERB
ajst-30200	128	14	variables	variable	NOUN
ajst-30200	128	15	such	such	ADJ
ajst-30200	128	16	as	as	ADP
ajst-30200	128	17	wave	wave	NOUN
ajst-30200	128	18	height	height	NOUN
ajst-30200	128	19	,	,	PUNCT
ajst-30200	128	20	frequency	frequency	NOUN
ajst-30200	128	21	,	,	PUNCT
ajst-30200	128	22	speed	speed	NOUN
ajst-30200	128	23	,	,	PUNCT
ajst-30200	128	24	and	and	CCONJ
ajst-30200	128	25	the	the	DET
ajst-30200	128	26	device	device	NOUN
ajst-30200	128	27	's	's	PART
ajst-30200	128	28	own	own	ADJ
ajst-30200	128	29	state	state	NOUN
ajst-30200	128	30	,	,	PUNCT
ajst-30200	128	31	such	such	ADJ
ajst-30200	128	32	as	as	ADP
ajst-30200	128	33	turbine	turbine	NOUN
ajst-30200	128	34	speed	speed	NOUN
ajst-30200	128	35	and	and	CCONJ
ajst-30200	128	36	water	water	NOUN
ajst-30200	128	37	column	column	NOUN
ajst-30200	128	38	oscillation	oscillation	NOUN
ajst-30200	128	39	amplitude	amplitude	NOUN
ajst-30200	128	40	.	.	PUNCT
ajst-30200	129	1	by	by	ADP
ajst-30200	129	2	monitoring	monitor	VERB
ajst-30200	129	3	these	these	DET
ajst-30200	129	4	state	state	NOUN
ajst-30200	129	5	variables	variable	NOUN
ajst-30200	129	6	,	,	PUNCT
ajst-30200	129	7	the	the	DET
ajst-30200	129	8	control	control	NOUN
ajst-30200	129	9	system	system	NOUN
ajst-30200	129	10	can	can	AUX
ajst-30200	129	11	be	be	AUX
ajst-30200	129	12	provided	provide	VERB
ajst-30200	129	13	with	with	ADP
ajst-30200	129	14	real	real	ADJ
ajst-30200	129	15	-	-	PUNCT
ajst-30200	129	16	time	time	NOUN
ajst-30200	129	17	environmental	environmental	ADJ
ajst-30200	129	18	data	datum	NOUN
ajst-30200	129	19	.	.	PUNCT
ajst-30200	130	1	deep	deep	ADJ
ajst-30200	130	2	neural	neural	ADJ
ajst-30200	130	3	networks	network	NOUN
ajst-30200	130	4	,	,	PUNCT
ajst-30200	130	5	such	such	ADJ
ajst-30200	130	6	as	as	ADP
ajst-30200	130	7	convolutional	convolutional	ADJ
ajst-30200	130	8	neural	neural	ADJ
ajst-30200	130	9	networks	network	NOUN
ajst-30200	130	10	(	(	PUNCT
ajst-30200	130	11	cnn	cnn	PROPN
ajst-30200	130	12	)	)	PUNCT
ajst-30200	130	13	or	or	CCONJ
ajst-30200	130	14	recurrent	recurrent	ADJ
ajst-30200	130	15	neural	neural	ADJ
ajst-30200	130	16	networks	network	NOUN
ajst-30200	130	17	(	(	PUNCT
ajst-30200	130	18	rnn	rnn	PROPN
ajst-30200	130	19	)	)	PUNCT
ajst-30200	130	20	,	,	PUNCT
ajst-30200	130	21	can	can	AUX
ajst-30200	130	22	be	be	AUX
ajst-30200	130	23	used	use	VERB
ajst-30200	130	24	for	for	ADP
ajst-30200	130	25	feature	feature	NOUN
ajst-30200	130	26	extraction	extraction	NOUN
ajst-30200	130	27	,	,	PUNCT
ajst-30200	130	28	transforming	transform	VERB
ajst-30200	130	29	raw	raw	ADJ
ajst-30200	130	30	environmental	environmental	ADJ
ajst-30200	130	31	data	datum	NOUN
ajst-30200	130	32	into	into	ADP
ajst-30200	130	33	meaningful	meaningful	ADJ
ajst-30200	130	34	high	high	ADJ
ajst-30200	130	35	-	-	PUNCT
ajst-30200	130	36	dimensional	dimensional	ADJ
ajst-30200	130	37	feature	feature	NOUN
ajst-30200	130	38	representations	representation	NOUN
ajst-30200	130	39	.	.	PUNCT
ajst-30200	131	1	the	the	DET
ajst-30200	131	2	goal	goal	NOUN
ajst-30200	131	3	of	of	ADP
ajst-30200	131	4	feature	feature	NOUN
ajst-30200	131	5	extraction	extraction	NOUN
ajst-30200	131	6	is	be	AUX
ajst-30200	131	7	to	to	PART
ajst-30200	131	8	convert	convert	VERB
ajst-30200	131	9	dynamic	dynamic	ADJ
ajst-30200	131	10	wave	wave	NOUN
ajst-30200	131	11	changes	change	NOUN
ajst-30200	131	12	into	into	ADP
ajst-30200	131	13	inputs	input	NOUN
ajst-30200	131	14	that	that	SCONJ
ajst-30200	131	15	the	the	DET
ajst-30200	131	16	system	system	NOUN
ajst-30200	131	17	can	can	AUX
ajst-30200	131	18	process	process	VERB
ajst-30200	131	19	,	,	PUNCT
ajst-30200	131	20	allowing	allow	VERB
ajst-30200	131	21	the	the	DET
ajst-30200	131	22	control	control	NOUN
ajst-30200	131	23	strategy	strategy	NOUN
ajst-30200	131	24	to	to	PART
ajst-30200	131	25	react	react	VERB
ajst-30200	131	26	quickly	quickly	ADV
ajst-30200	131	27	to	to	PART
ajst-30200	131	28	wave	wave	VERB
ajst-30200	131	29	variations	variation	NOUN
ajst-30200	131	30	.	.	PUNCT
ajst-30200	132	1	in	in	ADP
ajst-30200	132	2	complex	complex	ADJ
ajst-30200	132	3	environments	environment	NOUN
ajst-30200	132	4	,	,	PUNCT
ajst-30200	132	5	deep	deep	ADJ
ajst-30200	132	6	reinforcement	reinforcement	NOUN
ajst-30200	132	7	learning	learning	NOUN
ajst-30200	132	8	can	can	AUX
ajst-30200	132	9	automatically	automatically	ADV
ajst-30200	132	10	extract	extract	VERB
ajst-30200	132	11	the	the	DET
ajst-30200	132	12	key	key	ADJ
ajst-30200	132	13	factors	factor	NOUN
ajst-30200	132	14	influencing	influence	VERB
ajst-30200	132	15	energy	energy	NOUN
ajst-30200	132	16	conversion	conversion	NOUN
ajst-30200	132	17	efficiency	efficiency	NOUN
ajst-30200	132	18	without	without	ADP
ajst-30200	132	19	the	the	DET
ajst-30200	132	20	need	need	NOUN
ajst-30200	132	21	for	for	ADP
ajst-30200	132	22	manual	manual	ADJ
ajst-30200	132	23	feature	feature	NOUN
ajst-30200	132	24	design	design	NOUN
ajst-30200	132	25	.	.	PUNCT
ajst-30200	133	1	2	2	X
ajst-30200	133	2	.	.	X
ajst-30200	133	3	application	application	NOUN
ajst-30200	133	4	of	of	ADP
ajst-30200	133	5	reinforcement	reinforcement	NOUN
ajst-30200	133	6	learning	learning	NOUN
ajst-30200	133	7	algorithms	algorithm	NOUN
ajst-30200	133	8	in	in	ADP
ajst-30200	133	9	control	control	NOUN
ajst-30200	133	10	strategies	strategy	NOUN
ajst-30200	133	11	the	the	DET
ajst-30200	133	12	core	core	NOUN
ajst-30200	133	13	of	of	ADP
ajst-30200	133	14	reinforcement	reinforcement	NOUN
ajst-30200	133	15	learning	learning	NOUN
ajst-30200	133	16	algorithms	algorithm	NOUN
ajst-30200	133	17	is	be	AUX
ajst-30200	133	18	optimizing	optimize	VERB
ajst-30200	133	19	the	the	DET
ajst-30200	133	20	agent	agent	NOUN
ajst-30200	133	21	’s	’s	PART
ajst-30200	133	22	behavior	behavior	NOUN
ajst-30200	133	23	strategy	strategy	NOUN
ajst-30200	133	24	through	through	ADP
ajst-30200	133	25	a	a	DET
ajst-30200	133	26	reward	reward	NOUN
ajst-30200	133	27	mechanism	mechanism	NOUN
ajst-30200	133	28	.	.	PUNCT
ajst-30200	134	1	for	for	ADP
ajst-30200	134	2	the	the	DET
ajst-30200	134	3	control	control	NOUN
ajst-30200	134	4	of	of	ADP
ajst-30200	134	5	wave	wave	NOUN
ajst-30200	134	6	energy	energy	NOUN
ajst-30200	134	7	devices	device	NOUN
ajst-30200	134	8	,	,	PUNCT
ajst-30200	134	9	commonly	commonly	ADV
ajst-30200	134	10	used	use	VERB
ajst-30200	134	11	reinforcement	reinforcement	NOUN
ajst-30200	134	12	learning	learning	NOUN
ajst-30200	134	13	algorithms	algorithm	NOUN
ajst-30200	134	14	include	include	VERB
ajst-30200	134	15	deep	deep	ADJ
ajst-30200	134	16	q	q	NOUN
ajst-30200	134	17	networks	network	NOUN
ajst-30200	134	18	(	(	PUNCT
ajst-30200	134	19	dqn	dqn	PROPN
ajst-30200	134	20	)	)	PUNCT
ajst-30200	134	21	,	,	PUNCT
ajst-30200	134	22	policy	policy	NOUN
ajst-30200	134	23	gradient	gradient	NOUN
ajst-30200	134	24	methods	method	NOUN
ajst-30200	134	25	,	,	PUNCT
ajst-30200	134	26	and	and	CCONJ
ajst-30200	134	27	actor	actor	NOUN
ajst-30200	134	28	-	-	PUNCT
ajst-30200	134	29	critic	critic	NOUN
ajst-30200	134	30	methods	method	NOUN
ajst-30200	134	31	.	.	PUNCT
ajst-30200	135	1	deep	deep	ADJ
ajst-30200	135	2	q	q	NOUN
ajst-30200	135	3	network	network	NOUN
ajst-30200	135	4	(	(	PUNCT
ajst-30200	135	5	dqn	dqn	PROPN
ajst-30200	135	6	):	):	PUNCT
ajst-30200	135	7	dqn	dqn	NOUN
ajst-30200	135	8	approximates	approximate	VERB
ajst-30200	135	9	the	the	DET
ajst-30200	135	10	q	q	ADJ
ajst-30200	135	11	-	-	PUNCT
ajst-30200	135	12	value	value	NOUN
ajst-30200	135	13	function	function	NOUN
ajst-30200	135	14	using	use	VERB
ajst-30200	135	15	deep	deep	ADJ
ajst-30200	135	16	neural	neural	ADJ
ajst-30200	135	17	networks	network	NOUN
ajst-30200	135	18	to	to	PART
ajst-30200	135	19	evaluate	evaluate	VERB
ajst-30200	135	20	the	the	DET
ajst-30200	135	21	expected	expect	VERB
ajst-30200	135	22	reward	reward	NOUN
ajst-30200	135	23	for	for	ADP
ajst-30200	135	24	a	a	DET
ajst-30200	135	25	given	give	VERB
ajst-30200	135	26	action	action	NOUN
ajst-30200	135	27	in	in	ADP
ajst-30200	135	28	a	a	DET
ajst-30200	135	29	particular	particular	ADJ
ajst-30200	135	30	state	state	NOUN
ajst-30200	135	31	.	.	PUNCT
ajst-30200	136	1	by	by	ADP
ajst-30200	136	2	continually	continually	ADV
ajst-30200	136	3	updating	update	VERB
ajst-30200	136	4	the	the	DET
ajst-30200	136	5	q	q	ADJ
ajst-30200	136	6	-	-	PUNCT
ajst-30200	136	7	value	value	NOUN
ajst-30200	136	8	function	function	NOUN
ajst-30200	136	9	,	,	PUNCT
ajst-30200	136	10	the	the	DET
ajst-30200	136	11	agent	agent	NOUN
ajst-30200	136	12	learns	learn	VERB
ajst-30200	136	13	the	the	DET
ajst-30200	136	14	optimal	optimal	ADJ
ajst-30200	136	15	control	control	NOUN
ajst-30200	136	16	strategy	strategy	NOUN
ajst-30200	136	17	.	.	PUNCT
ajst-30200	137	1	the	the	DET
ajst-30200	137	2	advantage	advantage	NOUN
ajst-30200	137	3	of	of	ADP
ajst-30200	137	4	dqn	dqn	NOUN
ajst-30200	137	5	lies	lie	NOUN
ajst-30200	137	6	in	in	ADP
ajst-30200	137	7	its	its	PRON
ajst-30200	137	8	ability	ability	NOUN
ajst-30200	137	9	to	to	PART
ajst-30200	137	10	handle	handle	VERB
ajst-30200	137	11	high	high	ADJ
ajst-30200	137	12	-	-	PUNCT
ajst-30200	137	13	dimensional	dimensional	ADJ
ajst-30200	137	14	state	state	NOUN
ajst-30200	137	15	spaces	space	NOUN
ajst-30200	137	16	and	and	CCONJ
ajst-30200	137	17	efficiently	efficiently	ADV
ajst-30200	137	18	use	use	VERB
ajst-30200	137	19	past	past	ADJ
ajst-30200	137	20	experiences	experience	NOUN
ajst-30200	137	21	through	through	ADP
ajst-30200	137	22	experience	experience	NOUN
ajst-30200	137	23	replay	replay	NOUN
ajst-30200	137	24	during	during	ADP
ajst-30200	137	25	training	training	NOUN
ajst-30200	137	26	.	.	PUNCT
ajst-30200	138	1	policy	policy	NOUN
ajst-30200	138	2	gradient	gradient	NOUN
ajst-30200	138	3	methods	method	NOUN
ajst-30200	138	4	:	:	PUNCT
ajst-30200	138	5	policy	policy	NOUN
ajst-30200	138	6	gradient	gradient	NOUN
ajst-30200	138	7	methods	method	NOUN
ajst-30200	138	8	directly	directly	ADV
ajst-30200	138	9	optimize	optimize	VERB
ajst-30200	138	10	the	the	DET
ajst-30200	138	11	agent	agent	NOUN
ajst-30200	138	12	's	's	PART
ajst-30200	138	13	control	control	NOUN
ajst-30200	138	14	policy	policy	NOUN
ajst-30200	138	15	by	by	ADP
ajst-30200	138	16	calculating	calculate	VERB
ajst-30200	138	17	gradients	gradient	NOUN
ajst-30200	138	18	to	to	PART
ajst-30200	138	19	adjust	adjust	VERB
ajst-30200	138	20	policy	policy	NOUN
ajst-30200	138	21	parameters	parameter	NOUN
ajst-30200	138	22	.	.	PUNCT
ajst-30200	139	1	this	this	DET
ajst-30200	139	2	method	method	NOUN
ajst-30200	139	3	is	be	AUX
ajst-30200	139	4	suitable	suitable	ADJ
ajst-30200	139	5	for	for	ADP
ajst-30200	139	6	continuous	continuous	ADJ
ajst-30200	139	7	action	action	NOUN
ajst-30200	139	8	spaces	space	NOUN
ajst-30200	139	9	and	and	CCONJ
ajst-30200	139	10	can	can	AUX
ajst-30200	139	11	effectively	effectively	ADV
ajst-30200	139	12	optimize	optimize	VERB
ajst-30200	139	13	continuous	continuous	ADJ
ajst-30200	139	14	control	control	NOUN
ajst-30200	139	15	tasks	task	NOUN
ajst-30200	139	16	,	,	PUNCT
ajst-30200	139	17	such	such	ADJ
ajst-30200	139	18	as	as	ADP
ajst-30200	139	19	adjusting	adjust	VERB
ajst-30200	139	20	turbine	turbine	NOUN
ajst-30200	139	21	speed	speed	NOUN
ajst-30200	139	22	or	or	CCONJ
ajst-30200	139	23	controlling	control	VERB
ajst-30200	139	24	airflow	airflow	NOUN
ajst-30200	139	25	.	.	PUNCT
ajst-30200	140	1	actor	actor	NOUN
ajst-30200	140	2	-	-	PUNCT
ajst-30200	140	3	critic	critic	NOUN
ajst-30200	140	4	methods	method	NOUN
ajst-30200	140	5	:	:	PUNCT
ajst-30200	140	6	the	the	DET
ajst-30200	140	7	actor	actor	NOUN
ajst-30200	140	8	-	-	PUNCT
ajst-30200	140	9	critic	critic	NOUN
ajst-30200	140	10	method	method	NOUN
ajst-30200	140	11	combines	combine	VERB
ajst-30200	140	12	value	value	NOUN
ajst-30200	140	13	function	function	NOUN
ajst-30200	140	14	and	and	CCONJ
ajst-30200	140	15	policy	policy	NOUN
ajst-30200	140	16	function	function	NOUN
ajst-30200	140	17	,	,	PUNCT
ajst-30200	140	18	where	where	SCONJ
ajst-30200	140	19	the	the	DET
ajst-30200	140	20	critic	critic	NOUN
ajst-30200	140	21	evaluates	evaluate	VERB
ajst-30200	140	22	the	the	DET
ajst-30200	140	23	current	current	ADJ
ajst-30200	140	24	policy	policy	NOUN
ajst-30200	140	25	’s	’s	PART
ajst-30200	140	26	quality	quality	NOUN
ajst-30200	140	27	and	and	CCONJ
ajst-30200	140	28	the	the	DET
ajst-30200	140	29	actor	actor	NOUN
ajst-30200	140	30	adjusts	adjust	VERB
ajst-30200	140	31	its	its	PRON
ajst-30200	140	32	strategy	strategy	NOUN
ajst-30200	140	33	based	base	VERB
ajst-30200	140	34	on	on	ADP
ajst-30200	140	35	the	the	DET
ajst-30200	140	36	critic	critic	NOUN
ajst-30200	140	37	’s	’s	PART
ajst-30200	140	38	feedback	feedback	NOUN
ajst-30200	140	39	.	.	PUNCT
ajst-30200	141	1	this	this	DET
ajst-30200	141	2	method	method	NOUN
ajst-30200	141	3	demonstrates	demonstrate	VERB
ajst-30200	141	4	strong	strong	ADJ
ajst-30200	141	5	stability	stability	NOUN
ajst-30200	141	6	and	and	CCONJ
ajst-30200	141	7	training	training	NOUN
ajst-30200	141	8	efficiency	efficiency	NOUN
ajst-30200	141	9	,	,	PUNCT
ajst-30200	141	10	making	make	VERB
ajst-30200	141	11	it	it	PRON
ajst-30200	141	12	suitable	suitable	ADJ
ajst-30200	141	13	for	for	ADP
ajst-30200	141	14	complex	complex	ADJ
ajst-30200	141	15	control	control	NOUN
ajst-30200	141	16	tasks	task	NOUN
ajst-30200	141	17	.	.	PUNCT
ajst-30200	142	1	3	3	X
ajst-30200	142	2	.	.	X
ajst-30200	142	3	control	control	NOUN
ajst-30200	142	4	strategy	strategy	NOUN
ajst-30200	142	5	adjustment	adjustment	NOUN
ajst-30200	142	6	and	and	CCONJ
ajst-30200	142	7	optimization	optimization	NOUN
ajst-30200	142	8	the	the	DET
ajst-30200	142	9	adjustment	adjustment	NOUN
ajst-30200	142	10	and	and	CCONJ
ajst-30200	142	11	optimization	optimization	NOUN
ajst-30200	142	12	of	of	ADP
ajst-30200	142	13	control	control	NOUN
ajst-30200	142	14	strategies	strategy	NOUN
ajst-30200	142	15	are	be	AUX
ajst-30200	142	16	central	central	ADJ
ajst-30200	142	17	to	to	ADP
ajst-30200	142	18	the	the	DET
ajst-30200	142	19	adaptive	adaptive	ADJ
ajst-30200	142	20	control	control	NOUN
ajst-30200	142	21	system	system	NOUN
ajst-30200	142	22	based	base	VERB
ajst-30200	142	23	on	on	ADP
ajst-30200	142	24	deep	deep	ADJ
ajst-30200	142	25	reinforcement	reinforcement	NOUN
ajst-30200	142	26	learning	learning	NOUN
ajst-30200	142	27	.	.	PUNCT
ajst-30200	143	1	by	by	ADP
ajst-30200	143	2	analyzing	analyze	VERB
ajst-30200	143	3	feedback	feedback	NOUN
ajst-30200	143	4	from	from	ADP
ajst-30200	143	5	the	the	DET
ajst-30200	143	6	wave	wave	NOUN
ajst-30200	143	7	energy	energy	NOUN
ajst-30200	143	8	device	device	NOUN
ajst-30200	143	9	in	in	ADP
ajst-30200	143	10	real	real	ADJ
ajst-30200	143	11	-	-	PUNCT
ajst-30200	143	12	time	time	NOUN
ajst-30200	143	13	,	,	PUNCT
ajst-30200	143	14	the	the	DET
ajst-30200	143	15	control	control	NOUN
ajst-30200	143	16	system	system	NOUN
ajst-30200	143	17	can	can	AUX
ajst-30200	143	18	adjust	adjust	VERB
ajst-30200	143	19	the	the	DET
ajst-30200	143	20	turbine	turbine	NOUN
ajst-30200	143	21	working	work	VERB
ajst-30200	143	22	parameters	parameter	NOUN
ajst-30200	143	23	based	base	VERB
ajst-30200	143	24	on	on	ADP
ajst-30200	143	25	environmental	environmental	ADJ
ajst-30200	143	26	changes	change	NOUN
ajst-30200	143	27	.	.	PUNCT
ajst-30200	144	1	for	for	ADP
ajst-30200	144	2	example	example	NOUN
ajst-30200	144	3	,	,	PUNCT
ajst-30200	144	4	when	when	SCONJ
ajst-30200	144	5	wave	wave	NOUN
ajst-30200	144	6	frequency	frequency	NOUN
ajst-30200	144	7	is	be	AUX
ajst-30200	144	8	high	high	ADJ
ajst-30200	144	9	,	,	PUNCT
ajst-30200	144	10	the	the	DET
ajst-30200	144	11	system	system	NOUN
ajst-30200	144	12	can	can	AUX
ajst-30200	144	13	increase	increase	VERB
ajst-30200	144	14	the	the	DET
ajst-30200	144	15	turbine	turbine	NOUN
ajst-30200	144	16	speed	speed	NOUN
ajst-30200	144	17	;	;	PUNCT
ajst-30200	144	18	when	when	SCONJ
ajst-30200	144	19	the	the	DET
ajst-30200	144	20	wave	wave	NOUN
ajst-30200	144	21	frequency	frequency	NOUN
ajst-30200	144	22	is	be	AUX
ajst-30200	144	23	low	low	ADJ
ajst-30200	144	24	,	,	PUNCT
ajst-30200	144	25	it	it	PRON
ajst-30200	144	26	can	can	AUX
ajst-30200	144	27	reduce	reduce	VERB
ajst-30200	144	28	speed	speed	NOUN
ajst-30200	144	29	to	to	PART
ajst-30200	144	30	minimize	minimize	VERB
ajst-30200	144	31	energy	energy	NOUN
ajst-30200	144	32	loss	loss	NOUN
ajst-30200	144	33	.	.	PUNCT
ajst-30200	145	1	through	through	ADP
ajst-30200	145	2	deep	deep	ADJ
ajst-30200	145	3	reinforcement	reinforcement	NOUN
ajst-30200	145	4	learning	learning	NOUN
ajst-30200	145	5	,	,	PUNCT
ajst-30200	145	6	the	the	DET
ajst-30200	145	7	system	system	NOUN
ajst-30200	145	8	can	can	AUX
ajst-30200	145	9	automatically	automatically	ADV
ajst-30200	145	10	adjust	adjust	VERB
ajst-30200	145	11	control	control	NOUN
ajst-30200	145	12	strategies	strategy	NOUN
ajst-30200	145	13	under	under	ADP
ajst-30200	145	14	changing	change	VERB
ajst-30200	145	15	wave	wave	NOUN
ajst-30200	145	16	conditions	condition	NOUN
ajst-30200	145	17	,	,	PUNCT
ajst-30200	145	18	improving	improve	VERB
ajst-30200	145	19	energy	energy	NOUN
ajst-30200	145	20	conversion	conversion	NOUN
ajst-30200	145	21	efficiency	efficiency	NOUN
ajst-30200	145	22	.	.	PUNCT
ajst-30200	146	1	additionally	additionally	ADV
ajst-30200	146	2	,	,	PUNCT
ajst-30200	146	3	the	the	DET
ajst-30200	146	4	optimization	optimization	NOUN
ajst-30200	146	5	process	process	NOUN
ajst-30200	146	6	considers	consider	VERB
ajst-30200	146	7	not	not	PART
ajst-30200	146	8	only	only	ADV
ajst-30200	146	9	immediate	immediate	ADJ
ajst-30200	146	10	energy	energy	NOUN
ajst-30200	146	11	capture	capture	NOUN
ajst-30200	146	12	but	but	CCONJ
ajst-30200	146	13	also	also	ADV
ajst-30200	146	14	long	long	ADJ
ajst-30200	146	15	-	-	PUNCT
ajst-30200	146	16	term	term	NOUN
ajst-30200	146	17	operational	operational	ADJ
ajst-30200	146	18	stability	stability	NOUN
ajst-30200	146	19	.	.	PUNCT
ajst-30200	147	1	deep	deep	ADJ
ajst-30200	147	2	reinforcement	reinforcement	NOUN
ajst-30200	147	3	learning	learning	NOUN
ajst-30200	147	4	can	can	AUX
ajst-30200	147	5	incorporate	incorporate	VERB
ajst-30200	147	6	device	device	NOUN
ajst-30200	147	7	durability	durability	NOUN
ajst-30200	147	8	and	and	CCONJ
ajst-30200	147	9	stability	stability	NOUN
ajst-30200	147	10	into	into	ADP
ajst-30200	147	11	the	the	DET
ajst-30200	147	12	optimization	optimization	NOUN
ajst-30200	147	13	goal	goal	NOUN
ajst-30200	147	14	by	by	ADP
ajst-30200	147	15	designing	design	VERB
ajst-30200	147	16	appropriate	appropriate	ADJ
ajst-30200	147	17	reward	reward	NOUN
ajst-30200	147	18	functions	function	NOUN
ajst-30200	147	19	,	,	PUNCT
ajst-30200	147	20	ensuring	ensure	VERB
ajst-30200	147	21	the	the	DET
ajst-30200	147	22	control	control	NOUN
ajst-30200	147	23	strategy	strategy	NOUN
ajst-30200	147	24	maximizes	maximize	VERB
ajst-30200	147	25	energy	energy	NOUN
ajst-30200	147	26	output	output	NOUN
ajst-30200	147	27	while	while	SCONJ
ajst-30200	147	28	maintaining	maintain	VERB
ajst-30200	147	29	the	the	DET
ajst-30200	147	30	device	device	NOUN
ajst-30200	147	31	's	's	PART
ajst-30200	147	32	safety	safety	NOUN
ajst-30200	147	33	.	.	PUNCT
ajst-30200	148	1	(	(	PUNCT
ajst-30200	148	2	c	c	X
ajst-30200	148	3	)	)	PUNCT
ajst-30200	148	4	training	training	NOUN
ajst-30200	148	5	and	and	CCONJ
ajst-30200	148	6	optimization	optimization	NOUN
ajst-30200	148	7	of	of	ADP
ajst-30200	148	8	control	control	NOUN
ajst-30200	148	9	strategies	strategy	NOUN
ajst-30200	148	10	1	1	NUM
ajst-30200	148	11	.	.	PUNCT
ajst-30200	148	12	training	training	NOUN
ajst-30200	148	13	process	process	NOUN
ajst-30200	148	14	of	of	ADP
ajst-30200	148	15	reinforcement	reinforcement	NOUN
ajst-30200	148	16	learning	learning	NOUN
ajst-30200	148	17	models	model	VERB
ajst-30200	148	18	the	the	DET
ajst-30200	148	19	training	training	NOUN
ajst-30200	148	20	of	of	ADP
ajst-30200	148	21	reinforcement	reinforcement	NOUN
ajst-30200	148	22	learning	learning	NOUN
ajst-30200	148	23	models	model	NOUN
ajst-30200	148	24	typically	typically	ADV
ajst-30200	148	25	involves	involve	VERB
ajst-30200	148	26	the	the	DET
ajst-30200	148	27	following	follow	VERB
ajst-30200	148	28	steps	step	NOUN
ajst-30200	148	29	:	:	PUNCT
ajst-30200	148	30	environment	environment	NOUN
ajst-30200	148	31	initialization	initialization	NOUN
ajst-30200	148	32	,	,	PUNCT
ajst-30200	148	33	action	action	NOUN
ajst-30200	148	34	selection	selection	NOUN
ajst-30200	148	35	,	,	PUNCT
ajst-30200	148	36	reward	reward	VERB
ajst-30200	148	37	feedback	feedback	NOUN
ajst-30200	148	38	,	,	PUNCT
ajst-30200	148	39	and	and	CCONJ
ajst-30200	148	40	policy	policy	NOUN
ajst-30200	148	41	updates	update	NOUN
ajst-30200	148	42	.	.	PUNCT
ajst-30200	149	1	initially	initially	ADV
ajst-30200	149	2	,	,	PUNCT
ajst-30200	149	3	the	the	DET
ajst-30200	149	4	agent	agent	NOUN
ajst-30200	149	5	adopts	adopt	VERB
ajst-30200	149	6	a	a	DET
ajst-30200	149	7	random	random	ADJ
ajst-30200	149	8	control	control	NOUN
ajst-30200	149	9	strategy	strategy	NOUN
ajst-30200	149	10	and	and	CCONJ
ajst-30200	149	11	interacts	interact	VERB
ajst-30200	149	12	with	with	ADP
ajst-30200	149	13	the	the	DET
ajst-30200	149	14	environment	environment	NOUN
ajst-30200	149	15	.	.	PUNCT
ajst-30200	150	1	each	each	DET
ajst-30200	150	2	interaction	interaction	NOUN
ajst-30200	150	3	generates	generate	VERB
ajst-30200	150	4	a	a	DET
ajst-30200	150	5	combination	combination	NOUN
ajst-30200	150	6	of	of	ADP
ajst-30200	150	7	state	state	NOUN
ajst-30200	150	8	,	,	PUNCT
ajst-30200	150	9	action	action	NOUN
ajst-30200	150	10	,	,	PUNCT
ajst-30200	150	11	and	and	CCONJ
ajst-30200	150	12	reward	reward	NOUN
ajst-30200	150	13	,	,	PUNCT
ajst-30200	150	14	allowing	allow	VERB
ajst-30200	150	15	the	the	DET
ajst-30200	150	16	agent	agent	NOUN
ajst-30200	150	17	to	to	PART
ajst-30200	150	18	accumulate	accumulate	VERB
ajst-30200	150	19	experience	experience	NOUN
ajst-30200	150	20	and	and	CCONJ
ajst-30200	150	21	evaluate	evaluate	VERB
ajst-30200	150	22	the	the	DET
ajst-30200	150	23	effects	effect	NOUN
ajst-30200	150	24	of	of	ADP
ajst-30200	150	25	different	different	ADJ
ajst-30200	150	26	actions	action	NOUN
ajst-30200	150	27	.	.	PUNCT
ajst-30200	151	1	during	during	ADP
ajst-30200	151	2	training	training	NOUN
ajst-30200	151	3	,	,	PUNCT
ajst-30200	151	4	the	the	DET
ajst-30200	151	5	agent	agent	NOUN
ajst-30200	151	6	continually	continually	ADV
ajst-30200	151	7	adjusts	adjust	VERB
ajst-30200	151	8	its	its	PRON
ajst-30200	151	9	behavior	behavior	NOUN
ajst-30200	151	10	strategy	strategy	NOUN
ajst-30200	151	11	through	through	ADP
ajst-30200	151	12	q	q	ADJ
ajst-30200	151	13	-	-	PUNCT
ajst-30200	151	14	value	value	NOUN
ajst-30200	151	15	functions	function	NOUN
ajst-30200	151	16	,	,	PUNCT
ajst-30200	151	17	policy	policy	NOUN
ajst-30200	151	18	gradients	gradient	NOUN
ajst-30200	151	19	,	,	PUNCT
ajst-30200	151	20	or	or	CCONJ
ajst-30200	151	21	202	202	NUM
ajst-30200	151	22	actor	actor	NOUN
ajst-30200	151	23	-	-	PUNCT
ajst-30200	151	24	critic	critic	NOUN
ajst-30200	151	25	updates	update	NOUN
ajst-30200	151	26	.	.	PUNCT
ajst-30200	152	1	to	to	PART
ajst-30200	152	2	enhance	enhance	VERB
ajst-30200	152	3	training	training	NOUN
ajst-30200	152	4	stability	stability	NOUN
ajst-30200	152	5	and	and	CCONJ
ajst-30200	152	6	efficiency	efficiency	NOUN
ajst-30200	152	7	,	,	PUNCT
ajst-30200	152	8	experience	experience	NOUN
ajst-30200	152	9	replay	replay	NOUN
ajst-30200	152	10	is	be	AUX
ajst-30200	152	11	often	often	ADV
ajst-30200	152	12	used	use	VERB
ajst-30200	152	13	,	,	PUNCT
ajst-30200	152	14	where	where	SCONJ
ajst-30200	152	15	past	past	ADJ
ajst-30200	152	16	experiences	experience	NOUN
ajst-30200	152	17	are	be	AUX
ajst-30200	152	18	stored	store	VERB
ajst-30200	152	19	in	in	ADP
ajst-30200	152	20	a	a	DET
ajst-30200	152	21	memory	memory	NOUN
ajst-30200	152	22	pool	pool	NOUN
ajst-30200	152	23	and	and	CCONJ
ajst-30200	152	24	randomly	randomly	ADV
ajst-30200	152	25	sampled	sample	VERB
ajst-30200	152	26	for	for	ADP
ajst-30200	152	27	updates	update	NOUN
ajst-30200	152	28	,	,	PUNCT
ajst-30200	152	29	breaking	break	VERB
ajst-30200	152	30	data	datum	NOUN
ajst-30200	152	31	correlations	correlation	NOUN
ajst-30200	152	32	and	and	CCONJ
ajst-30200	152	33	avoiding	avoid	VERB
ajst-30200	152	34	overfitting	overfitte	VERB
ajst-30200	152	35	.	.	PUNCT
ajst-30200	153	1	2	2	X
ajst-30200	153	2	.	.	X
ajst-30200	153	3	design	design	NOUN
ajst-30200	153	4	and	and	CCONJ
ajst-30200	153	5	adjustment	adjustment	NOUN
ajst-30200	153	6	of	of	ADP
ajst-30200	153	7	reward	reward	NOUN
ajst-30200	153	8	functions	function	NOUN
ajst-30200	153	9	the	the	DET
ajst-30200	153	10	reward	reward	NOUN
ajst-30200	153	11	function	function	NOUN
ajst-30200	153	12	is	be	AUX
ajst-30200	153	13	crucial	crucial	ADJ
ajst-30200	153	14	in	in	ADP
ajst-30200	153	15	determining	determine	VERB
ajst-30200	153	16	the	the	DET
ajst-30200	153	17	agent	agent	NOUN
ajst-30200	153	18	’s	’s	PART
ajst-30200	153	19	behavior	behavior	NOUN
ajst-30200	153	20	.	.	PUNCT
ajst-30200	154	1	designing	design	VERB
ajst-30200	154	2	a	a	DET
ajst-30200	154	3	reasonable	reasonable	ADJ
ajst-30200	154	4	reward	reward	NOUN
ajst-30200	154	5	function	function	NOUN
ajst-30200	154	6	helps	help	VERB
ajst-30200	154	7	guide	guide	VERB
ajst-30200	154	8	the	the	DET
ajst-30200	154	9	agent	agent	NOUN
ajst-30200	154	10	toward	toward	ADP
ajst-30200	154	11	optimizing	optimize	VERB
ajst-30200	154	12	the	the	DET
ajst-30200	154	13	control	control	NOUN
ajst-30200	154	14	strategy	strategy	NOUN
ajst-30200	154	15	.	.	PUNCT
ajst-30200	155	1	in	in	ADP
ajst-30200	155	2	the	the	DET
ajst-30200	155	3	control	control	NOUN
ajst-30200	155	4	of	of	ADP
ajst-30200	155	5	wave	wave	NOUN
ajst-30200	155	6	energy	energy	NOUN
ajst-30200	155	7	devices	device	NOUN
ajst-30200	155	8	,	,	PUNCT
ajst-30200	155	9	the	the	DET
ajst-30200	155	10	reward	reward	NOUN
ajst-30200	155	11	function	function	NOUN
ajst-30200	155	12	must	must	AUX
ajst-30200	155	13	account	account	VERB
ajst-30200	155	14	for	for	ADP
ajst-30200	155	15	factors	factor	NOUN
ajst-30200	155	16	like	like	ADP
ajst-30200	155	17	energy	energy	NOUN
ajst-30200	155	18	conversion	conversion	NOUN
ajst-30200	155	19	efficiency	efficiency	NOUN
ajst-30200	155	20	,	,	PUNCT
ajst-30200	155	21	system	system	NOUN
ajst-30200	155	22	stability	stability	NOUN
ajst-30200	155	23	,	,	PUNCT
ajst-30200	155	24	and	and	CCONJ
ajst-30200	155	25	device	device	NOUN
ajst-30200	155	26	longevity	longevity	NOUN
ajst-30200	155	27	.	.	PUNCT
ajst-30200	156	1	for	for	ADP
ajst-30200	156	2	example	example	NOUN
ajst-30200	156	3	,	,	PUNCT
ajst-30200	156	4	the	the	DET
ajst-30200	156	5	reward	reward	NOUN
ajst-30200	156	6	can	can	AUX
ajst-30200	156	7	be	be	AUX
ajst-30200	156	8	set	set	VERB
ajst-30200	156	9	based	base	VERB
ajst-30200	156	10	on	on	ADP
ajst-30200	156	11	the	the	DET
ajst-30200	156	12	turbine	turbine	NOUN
ajst-30200	156	13	's	's	PART
ajst-30200	156	14	output	output	NOUN
ajst-30200	156	15	power	power	NOUN
ajst-30200	156	16	,	,	PUNCT
ajst-30200	156	17	with	with	ADP
ajst-30200	156	18	penalty	penalty	NOUN
ajst-30200	156	19	terms	term	NOUN
ajst-30200	156	20	related	relate	VERB
ajst-30200	156	21	to	to	ADP
ajst-30200	156	22	the	the	DET
ajst-30200	156	23	device	device	NOUN
ajst-30200	156	24	's	's	PART
ajst-30200	156	25	health	health	NOUN
ajst-30200	156	26	status	status	NOUN
ajst-30200	156	27	to	to	PART
ajst-30200	156	28	prevent	prevent	VERB
ajst-30200	156	29	excessive	excessive	ADJ
ajst-30200	156	30	oscillations	oscillation	NOUN
ajst-30200	156	31	or	or	CCONJ
ajst-30200	156	32	structural	structural	ADJ
ajst-30200	156	33	damage	damage	NOUN
ajst-30200	156	34	.	.	PUNCT
ajst-30200	157	1	adjusting	adjust	VERB
ajst-30200	157	2	the	the	DET
ajst-30200	157	3	reward	reward	NOUN
ajst-30200	157	4	function	function	NOUN
ajst-30200	157	5	is	be	AUX
ajst-30200	157	6	also	also	ADV
ajst-30200	157	7	a	a	DET
ajst-30200	157	8	critical	critical	ADJ
ajst-30200	157	9	optimization	optimization	NOUN
ajst-30200	157	10	process	process	NOUN
ajst-30200	157	11	.	.	PUNCT
ajst-30200	158	1	during	during	ADP
ajst-30200	158	2	training	training	NOUN
ajst-30200	158	3	,	,	PUNCT
ajst-30200	158	4	the	the	DET
ajst-30200	158	5	reward	reward	NOUN
ajst-30200	158	6	function	function	NOUN
ajst-30200	158	7	can	can	AUX
ajst-30200	158	8	be	be	AUX
ajst-30200	158	9	dynamically	dynamically	ADV
ajst-30200	158	10	adjusted	adjust	VERB
ajst-30200	158	11	based	base	VERB
ajst-30200	158	12	on	on	ADP
ajst-30200	158	13	the	the	DET
ajst-30200	158	14	agent	agent	NOUN
ajst-30200	158	15	's	's	PART
ajst-30200	158	16	performance	performance	NOUN
ajst-30200	158	17	to	to	PART
ajst-30200	158	18	ensure	ensure	VERB
ajst-30200	158	19	that	that	SCONJ
ajst-30200	158	20	optimization	optimization	NOUN
ajst-30200	158	21	goals	goal	NOUN
ajst-30200	158	22	are	be	AUX
ajst-30200	158	23	effectively	effectively	ADV
ajst-30200	158	24	achieved	achieve	VERB
ajst-30200	158	25	.	.	PUNCT
ajst-30200	159	1	3	3	X
ajst-30200	159	2	.	.	X
ajst-30200	159	3	evaluation	evaluation	NOUN
ajst-30200	159	4	and	and	CCONJ
ajst-30200	159	5	adjustment	adjustment	NOUN
ajst-30200	159	6	of	of	ADP
ajst-30200	159	7	control	control	NOUN
ajst-30200	159	8	effectiveness	effectiveness	NOUN
ajst-30200	159	9	the	the	DET
ajst-30200	159	10	effectiveness	effectiveness	NOUN
ajst-30200	159	11	of	of	ADP
ajst-30200	159	12	the	the	DET
ajst-30200	159	13	control	control	NOUN
ajst-30200	159	14	strategy	strategy	NOUN
ajst-30200	159	15	needs	need	VERB
ajst-30200	159	16	to	to	PART
ajst-30200	159	17	be	be	AUX
ajst-30200	159	18	evaluated	evaluate	VERB
ajst-30200	159	19	through	through	ADP
ajst-30200	159	20	simulation	simulation	NOUN
ajst-30200	159	21	and	and	CCONJ
ajst-30200	159	22	experimental	experimental	ADJ
ajst-30200	159	23	verification	verification	NOUN
ajst-30200	159	24	.	.	PUNCT
ajst-30200	160	1	in	in	ADP
ajst-30200	160	2	the	the	DET
ajst-30200	160	3	practical	practical	ADJ
ajst-30200	160	4	application	application	NOUN
ajst-30200	160	5	of	of	ADP
ajst-30200	160	6	wave	wave	NOUN
ajst-30200	160	7	energy	energy	NOUN
ajst-30200	160	8	devices	device	NOUN
ajst-30200	160	9	,	,	PUNCT
ajst-30200	160	10	comparing	compare	VERB
ajst-30200	160	11	the	the	DET
ajst-30200	160	12	performance	performance	NOUN
ajst-30200	160	13	of	of	ADP
ajst-30200	160	14	traditional	traditional	ADJ
ajst-30200	160	15	control	control	NOUN
ajst-30200	160	16	methods	method	NOUN
ajst-30200	160	17	with	with	ADP
ajst-30200	160	18	deep	deep	ADJ
ajst-30200	160	19	reinforcement	reinforcement	NOUN
ajst-30200	160	20	learning	learning	NOUN
ajst-30200	160	21	methods	method	NOUN
ajst-30200	160	22	can	can	AUX
ajst-30200	160	23	assess	assess	VERB
ajst-30200	160	24	the	the	DET
ajst-30200	160	25	advantages	advantage	NOUN
ajst-30200	160	26	of	of	ADP
ajst-30200	160	27	drl	drl	PROPN
ajst-30200	160	28	control	control	NOUN
ajst-30200	160	29	strategies	strategy	NOUN
ajst-30200	160	30	.	.	PUNCT
ajst-30200	161	1	evaluation	evaluation	NOUN
ajst-30200	161	2	metrics	metric	NOUN
ajst-30200	161	3	include	include	VERB
ajst-30200	161	4	energy	energy	NOUN
ajst-30200	161	5	conversion	conversion	NOUN
ajst-30200	161	6	efficiency	efficiency	NOUN
ajst-30200	161	7	,	,	PUNCT
ajst-30200	161	8	device	device	NOUN
ajst-30200	161	9	stability	stability	NOUN
ajst-30200	161	10	,	,	PUNCT
ajst-30200	161	11	and	and	CCONJ
ajst-30200	161	12	turbine	turbine	NOUN
ajst-30200	161	13	operational	operational	ADJ
ajst-30200	161	14	status	status	NOUN
ajst-30200	161	15	.	.	PUNCT
ajst-30200	162	1	through	through	ADP
ajst-30200	162	2	multiple	multiple	ADJ
ajst-30200	162	3	rounds	round	NOUN
ajst-30200	162	4	of	of	ADP
ajst-30200	162	5	training	training	NOUN
ajst-30200	162	6	and	and	CCONJ
ajst-30200	162	7	evaluation	evaluation	NOUN
ajst-30200	162	8	,	,	PUNCT
ajst-30200	162	9	the	the	DET
ajst-30200	162	10	agent	agent	NOUN
ajst-30200	162	11	can	can	AUX
ajst-30200	162	12	gradually	gradually	ADV
ajst-30200	162	13	optimize	optimize	VERB
ajst-30200	162	14	the	the	DET
ajst-30200	162	15	control	control	NOUN
ajst-30200	162	16	strategy	strategy	NOUN
ajst-30200	162	17	under	under	ADP
ajst-30200	162	18	various	various	ADJ
ajst-30200	162	19	wave	wave	NOUN
ajst-30200	162	20	conditions	condition	NOUN
ajst-30200	162	21	.	.	PUNCT
ajst-30200	163	1	ultimately	ultimately	ADV
ajst-30200	163	2	,	,	PUNCT
ajst-30200	163	3	the	the	DET
ajst-30200	163	4	adaptive	adaptive	ADJ
ajst-30200	163	5	control	control	NOUN
ajst-30200	163	6	strategy	strategy	NOUN
ajst-30200	163	7	based	base	VERB
ajst-30200	163	8	on	on	ADP
ajst-30200	163	9	deep	deep	ADJ
ajst-30200	163	10	reinforcement	reinforcement	NOUN
ajst-30200	163	11	learning	learning	NOUN
ajst-30200	163	12	will	will	AUX
ajst-30200	163	13	enable	enable	VERB
ajst-30200	163	14	efficient	efficient	ADJ
ajst-30200	163	15	operation	operation	NOUN
ajst-30200	163	16	of	of	ADP
ajst-30200	163	17	the	the	DET
ajst-30200	163	18	wave	wave	NOUN
ajst-30200	163	19	energy	energy	NOUN
ajst-30200	163	20	device	device	NOUN
ajst-30200	163	21	while	while	SCONJ
ajst-30200	163	22	ensuring	ensure	VERB
ajst-30200	163	23	its	its	PRON
ajst-30200	163	24	stability	stability	NOUN
ajst-30200	163	25	and	and	CCONJ
ajst-30200	163	26	long	long	ADJ
ajst-30200	163	27	-	-	PUNCT
ajst-30200	163	28	term	term	NOUN
ajst-30200	163	29	reliability	reliability	NOUN
ajst-30200	163	30	in	in	ADP
ajst-30200	163	31	complex	complex	ADJ
ajst-30200	163	32	environments	environment	NOUN
ajst-30200	163	33	.	.	PUNCT
ajst-30200	164	1	4	4	X
ajst-30200	164	2	.	.	X
ajst-30200	164	3	conclusion	conclusion	VERB
ajst-30200	164	4	the	the	DET
ajst-30200	164	5	adaptive	adaptive	ADJ
ajst-30200	164	6	control	control	NOUN
ajst-30200	164	7	strategy	strategy	NOUN
ajst-30200	164	8	based	base	VERB
ajst-30200	164	9	on	on	ADP
ajst-30200	164	10	deep	deep	ADJ
ajst-30200	164	11	reinforcement	reinforcement	NOUN
ajst-30200	164	12	learning	learning	NOUN
ajst-30200	164	13	significantly	significantly	ADV
ajst-30200	164	14	improves	improve	VERB
ajst-30200	164	15	the	the	DET
ajst-30200	164	16	energy	energy	NOUN
ajst-30200	164	17	conversion	conversion	NOUN
ajst-30200	164	18	efficiency	efficiency	NOUN
ajst-30200	164	19	and	and	CCONJ
ajst-30200	164	20	stability	stability	NOUN
ajst-30200	164	21	of	of	ADP
ajst-30200	164	22	the	the	DET
ajst-30200	164	23	oscillating	oscillate	VERB
ajst-30200	164	24	water	water	NOUN
ajst-30200	164	25	column	column	NOUN
ajst-30200	164	26	wave	wave	VERB
ajst-30200	164	27	energy	energy	NOUN
ajst-30200	164	28	device	device	NOUN
ajst-30200	164	29	.	.	PUNCT
ajst-30200	165	1	by	by	ADP
ajst-30200	165	2	adjusting	adjust	VERB
ajst-30200	165	3	turbine	turbine	NOUN
ajst-30200	165	4	control	control	NOUN
ajst-30200	165	5	parameters	parameter	NOUN
ajst-30200	165	6	in	in	ADP
ajst-30200	165	7	real	real	ADJ
ajst-30200	165	8	time	time	NOUN
ajst-30200	165	9	,	,	PUNCT
ajst-30200	165	10	this	this	DET
ajst-30200	165	11	method	method	NOUN
ajst-30200	165	12	effectively	effectively	ADV
ajst-30200	165	13	addresses	address	VERB
ajst-30200	165	14	the	the	DET
ajst-30200	165	15	dynamic	dynamic	ADJ
ajst-30200	165	16	changes	change	NOUN
ajst-30200	165	17	in	in	ADP
ajst-30200	165	18	wave	wave	NOUN
ajst-30200	165	19	conditions	condition	NOUN
ajst-30200	165	20	and	and	CCONJ
ajst-30200	165	21	optimizes	optimize	VERB
ajst-30200	165	22	the	the	DET
ajst-30200	165	23	device	device	NOUN
ajst-30200	165	24	’s	’s	PART
ajst-30200	165	25	performance	performance	NOUN
ajst-30200	165	26	over	over	ADP
ajst-30200	165	27	long	long	ADJ
ajst-30200	165	28	-	-	PUNCT
ajst-30200	165	29	term	term	NOUN
ajst-30200	165	30	operation	operation	NOUN
ajst-30200	165	31	,	,	PUNCT
ajst-30200	165	32	showing	show	VERB
ajst-30200	165	33	broad	broad	ADJ
ajst-30200	165	34	application	application	NOUN
ajst-30200	165	35	prospects	prospect	NOUN
ajst-30200	165	36	.	.	PUNCT
ajst-30200	166	1	references	reference	NOUN
ajst-30200	166	2	[	[	X
ajst-30200	166	3	1	1	X
ajst-30200	166	4	]	]	PUNCT
ajst-30200	166	5	gonçalves	gonçalve	VERB
ajst-30200	166	6	r	r	NOUN
ajst-30200	166	7	a	a	DET
ajst-30200	166	8	a	a	DET
ajst-30200	166	9	c	c	NOUN
ajst-30200	166	10	,	,	PUNCT
ajst-30200	166	11	teixeira	teixeira	PROPN
ajst-30200	166	12	p	p	NOUN
ajst-30200	166	13	r	r	PROPN
ajst-30200	166	14	f	f	PROPN
ajst-30200	166	15	,	,	PUNCT
ajst-30200	166	16	didier	didier	PROPN
ajst-30200	166	17	e	e	PROPN
ajst-30200	166	18	,	,	PUNCT
ajst-30200	166	19	et	et	NOUN
ajst-30200	166	20	al.numerical	al.numerical	NOUN
ajst-30200	166	21	analysis	analysis	NOUN
ajst-30200	166	22	of	of	ADP
ajst-30200	166	23	the	the	DET
ajst-30200	166	24	influence	influence	NOUN
ajst-30200	166	25	of	of	ADP
ajst-30200	166	26	air	air	NOUN
ajst-30200	166	27	compressibility	compressibility	NOUN
ajst-30200	166	28	effects	effect	NOUN
ajst-30200	166	29	on	on	ADP
ajst-30200	166	30	an	an	DET
ajst-30200	166	31	oscillating	oscillate	VERB
ajst-30200	166	32	water	water	NOUN
ajst-30200	166	33	column	column	NOUN
ajst-30200	166	34	wave	wave	VERB
ajst-30200	166	35	energy	energy	NOUN
ajst-30200	166	36	converter	converter	NOUN
ajst-30200	166	37	chamber	chamber	NOUN
ajst-30200	167	1	[	[	X
ajst-30200	167	2	j	j	X
ajst-30200	167	3	]	]	X
ajst-30200	167	4	.	.	PUNCT
ajst-30200	168	1	renewable	renewable	ADJ
ajst-30200	168	2	energy	energy	NOUN
ajst-30200	168	3	,	,	PUNCT
ajst-30200	168	4	2020	2020	NUM
ajst-30200	168	5	,	,	PUNCT
ajst-30200	168	6	153	153	NUM
ajst-30200	168	7	:	:	SYM
ajst-30200	168	8	1183	1183	NUM
ajst-30200	168	9	-	-	SYM
ajst-30200	168	10	1193	1193	NUM
ajst-30200	168	11	.	.	PUNCT
ajst-30200	169	1	doi	doi	NOUN
ajst-30200	169	2	:	:	PUNCT
ajst-30200	169	3	10.1016	10.1016	NUM
ajst-30200	169	4	/	/	SYM
ajst-30200	169	5	j.renene.2020.02.080	j.renene.2020.02.080	PROPN
ajst-30200	169	6	.	.	PUNCT
ajst-30200	170	1	[	[	X
ajst-30200	170	2	2	2	NUM
ajst-30200	170	3	]	]	PUNCT
ajst-30200	170	4	cui	cui	NOUN
ajst-30200	170	5	y	y	PROPN
ajst-30200	170	6	,	,	PUNCT
ajst-30200	170	7	liu	liu	PROPN
ajst-30200	170	8	z	z	PROPN
ajst-30200	170	9	,	,	PUNCT
ajst-30200	170	10	zhang	zhang	PROPN
ajst-30200	170	11	x	x	SYM
ajst-30200	170	12	x	x	X
ajst-30200	170	13	,	,	PUNCT
ajst-30200	170	14	et	et	PROPN
ajst-30200	170	15	al.self	al.self	PRON
ajst-30200	170	16	-	-	PUNCT
ajst-30200	170	17	starting	start	VERB
ajst-30200	170	18	analysis	analysis	NOUN
ajst-30200	170	19	of	of	ADP
ajst-30200	170	20	an	an	DET
ajst-30200	170	21	owc	owc	PROPN
ajst-30200	170	22	axial	axial	ADJ
ajst-30200	170	23	impulse	impulse	ADJ
ajst-30200	170	24	turbine	turbine	NOUN
ajst-30200	170	25	in	in	ADP
ajst-30200	170	26	constant	constant	ADJ
ajst-30200	170	27	flows	flow	NOUN
ajst-30200	170	28	:	:	PUNCT
ajst-30200	170	29	experimental	experimental	ADJ
ajst-30200	170	30	and	and	CCONJ
ajst-30200	170	31	numerical	numerical	ADJ
ajst-30200	170	32	studies[j].applied	studies[j].applie	VERB
ajst-30200	170	33	ocean	ocean	PROPN
ajst-30200	170	34	research,2019,82	research,2019,82	NOUN
ajst-30200	170	35	:	:	PUNCT
ajst-30200	170	36	458469.doi	458469.doi	NUM
ajst-30200	170	37	:	:	PUNCT
ajst-30200	170	38	10.1016	10.1016	NUM
ajst-30200	170	39	/	/	SYM
ajst-30200	170	40	j.apor.2018.11.014	j.apor.2018.11.014	NOUN
ajst-30200	170	41	.	.	PUNCT
ajst-30200	171	1	[	[	X
ajst-30200	171	2	3	3	X
ajst-30200	171	3	]	]	X
ajst-30200	171	4	liu	liu	PROPN
ajst-30200	171	5	z	z	PROPN
ajst-30200	171	6	,	,	PUNCT
ajst-30200	171	7	jin	jin	PROPN
ajst-30200	171	8	j	j	PROPN
ajst-30200	171	9	y	y	PROPN
ajst-30200	171	10	,	,	PUNCT
ajst-30200	171	11	cui	cui	VERB
ajst-30200	171	12	y	y	PROPN
ajst-30200	171	13	,	,	PUNCT
ajst-30200	171	14	et	et	NOUN
ajst-30200	171	15	al.numerical	al.numerical	NOUN
ajst-30200	171	16	analysis	analysis	NOUN
ajst-30200	171	17	of	of	ADP
ajst-30200	171	18	impulse	impulse	ADJ
ajst-30200	171	19	turbine	turbine	NOUN
ajst-30200	171	20	for	for	ADP
ajst-30200	171	21	isolated	isolated	ADJ
ajst-30200	171	22	pilot	pilot	NOUN
ajst-30200	171	23	owc	owc	NOUN
ajst-30200	171	24	system	system	NOUN
ajst-30200	172	1	[	[	X
ajst-30200	172	2	j	j	X
ajst-30200	172	3	]	]	X
ajst-30200	172	4	.	.	PUNCT
ajst-30200	173	1	advances	advance	NOUN
ajst-30200	173	2	in	in	ADP
ajst-30200	173	3	mechanical	mechanical	ADJ
ajst-30200	173	4	engineering	engineering	NOUN
ajst-30200	173	5	,	,	PUNCT
ajst-30200	173	6	2013.doi	2013.doi	NUM
ajst-30200	173	7	:	:	PUNCT
ajst-30200	173	8	10.1155/	10.1155/	NUM
ajst-30200	173	9	2013/416109	2013/416109	NUM
ajst-30200	173	10	.	.	PUNCT
ajst-30200	174	1	[	[	X
ajst-30200	174	2	4	4	X
ajst-30200	174	3	]	]	X
ajst-30200	174	4	liu	liu	PROPN
ajst-30200	174	5	z	z	PROPN
ajst-30200	174	6	,	,	PUNCT
ajst-30200	174	7	xu	xu	PROPN
ajst-30200	174	8	c	c	PROPN
ajst-30200	174	9	l	l	PROPN
ajst-30200	174	10	,	,	PUNCT
ajst-30200	174	11	shi	shi	PROPN
ajst-30200	174	12	h	h	PROPN
ajst-30200	175	1	d	d	PROPN
ajst-30200	175	2	,	,	PUNCT
ajst-30200	175	3	et	et	NOUN
ajst-30200	175	4	al.wave	al.wave	NOUN
ajst-30200	175	5	-	-	NOUN
ajst-30200	175	6	flume	flume	ADJ
ajst-30200	175	7	tests	test	NOUN
ajst-30200	175	8	of	of	ADP
ajst-30200	175	9	a	a	DET
ajst-30200	175	10	modelscaled	modelscaled	ADJ
ajst-30200	175	11	owc	owc	PROPN
ajst-30200	175	12	chamber	chamber	NOUN
ajst-30200	175	13	-	-	PUNCT
ajst-30200	175	14	turbine	turbine	NOUN
ajst-30200	175	15	system	system	NOUN
ajst-30200	175	16	under	under	ADP
ajst-30200	175	17	irregular	irregular	ADJ
ajst-30200	175	18	wave	wave	NOUN
ajst-30200	175	19	conditions	condition	NOUN
ajst-30200	176	1	[	[	X
ajst-30200	176	2	j	j	X
ajst-30200	176	3	]	]	X
ajst-30200	176	4	.	.	PUNCT
ajst-30200	177	1	applied	applied	PROPN
ajst-30200	177	2	ocean	ocean	PROPN
ajst-30200	177	3	research,2020,99	research,2020,99	PROPN
ajst-30200	177	4	:	:	PUNCT
ajst-30200	177	5	102141.doi	102141.doi	NUM
ajst-30200	177	6	:	:	PUNCT
ajst-30200	177	7	10.1016	10.1016	NUM
ajst-30200	177	8	/	/	SYM
ajst-30200	177	9	j.apor.2020.102141	j.apor.2020.102141	ADJ
ajst-30200	177	10	.	.	PUNCT