Sebuah Kajian Pustaka: JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 ISSN 2541-6332 | e-ISSN 2548-4281 Journal homepage: http://ejournal.umm.ac.id/index.php/JEMMME Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 209 Online Blade Washing Analysis on Gas Turbine Performance in Power Plants Wilarsoa, Ari Dwi Wibowob a,b Sekolah Tinggi Teknologi Muhammadiyah Cileungsi, Bogor. Jl. Anggrek, No. 25, Komplek PTSC-Cileungsi, Bogor, Jawa Barat-Indonesia 021-82495502 e-mail: wilarso@sttmcileungsi, aridwiwibowo71@gmail.com Abstract The main problem that often occurs in the operation and maintenance of power plants is a decrease in the reliability of the gas turbine. The decline in the performance of the gas turbine, which often experiences trips, was recorded at the highest 3 times in one day. Based on the inspection, it was found that there were deposits on the compressor and turbine blades during operation. The decrease in power in the generating unit is accompanied by an increase in fuel consumption. The purpose of this study is to analyze blade washing online on the performance of gas turbines due to the formation of carbon deposits on the compressor wheel and turbine wheel. To improve the reliability of the gas engine, a method of doing blade washing is needed to clean carbon deposits in the compressor and turbine wheel. Based on the results of research before blade washing the turbine power only reached 255.37621 MW, after blade washing was able to make the compressor work more reliably, produce good turbine gas efficiency, and be able to reduce turbine gas performance disturbances due to running hours the power generated reached 268,77738 MW, there is a fuel consumption savings of 1.4 kg/s and thermal efficiency of 0.8%. Online washing is carried out at a load condition of 200MW ±5MW. To clean fouling and maintain the performance of the turbine. Cleanliness of the compressor and turbine blades can be maintained by carrying out this blade washing based on a periodic schedule calculated based on running hours. Keywords: blade washing on-line; compressor & turbine wheel; turbine gas efficiency 1. INTRODUCTION Gas-Fired Power Plant Extension Project (GFPPEP) with a capacity of 740 MW. The power plant has a combined cycle scheme consisting of two Gas Turbines (GT) namely GT 3.1 & GT 3.2 with type and has two Heat Recovery Steam Generators (HRSG) and one Steam Turbine [1]. This GT uses natural gas fuel and produces 235 MW (at installed power) for each unit, while for backup if natural gas cannot be supplied by PT. X, it uses high- speed diesel (HSD). PLTGU is an equipment installation that functions to convert heat energy (the result of burning fuel and air) into useful electrical energy. This PLTGU system is a combination of PLTG and PLTU [2]. PLTU utilizes heat energy and steam from exhaust gases resulting from combustion in the PLTG to heat water in the HRSG (Heat Recovery Steam Generator), so that it becomes dry saturated steam. This dry saturated steam will be used to turn the turbine blades [2]. The gas produced in the combustion chamber at the gas power plant [3] then moves the turbine blades mechanically and because of the location of the generator on one shaft with the turbine it will drive the generator [2], This mechanism will convert it into electrical energy. Similar to PLTU, PLTG fuel can be in the form of liquid http://ejournal.umm.ac.id/index.php/JEMMME JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 210 (BBM) or gas (natural gas). The use of fuel determines the level of combust ion efficiency and the process [2][4]. Gas turbine generators often experience trips up to 3 times a day, this will affect the productivity of the generator. The factor of decreasing the performance of the gas turbine, due to the presence of a deposit on the surface of the compressor and turbine blades [5]. Deposits attached to the compressor blades can reduce the supply of air to the combustion chamber, and can hamper the overall performance of the gas turbine [6]. The formation of deposits on the compressor and turbine blades is due to the imperfect air to fuel ratio [7] [8]. The process of deposit formation is a reaction between air-containing dust and fuel in the combustor chamber [9][10]. Polluted air contains dust, sand, hydrocarbon vapors, insects, and salt. Figure 1 describes the scheme of the generation process from the PLTU/PLTGU. Figure 1. Gas and steam power plant process To keep the blade compressor performance clean, it must be cleaned online or off- line, in other words, it can be carried out on the condition of the gas turbine being loaded or unloaded. The method is carried out by spraying pressurized water into the compressor blade, to reduce deposits on the blade compressor surface [11]. With the physical condition of the working environment around the gas turbine which has a risk of air contamination of 180 μg/Nm3 (around the power plant area there is loading and unloading of sand), coal storage, and cooking oil factory) further studies need to be done to find out how much effectiveness cleaning the compressor and turbine blades have on the overall gas turbine efficiency [12]. To be a reference in finding the most optimal time in carrying out this online blade washing method [13][14]. 2. METHODS In this research, unit performance data is needed to carry out the calculation process. The data is taken and obtained from the results of observations and recordings stored in the computer system while the unit is operating [15]. 1) Compressor efficiency after cleaning blade washing on-line, 2) Gas turbine efficiency after cleaning by on-line blade washing, 3) Fuel efficiency after cleaning by on-line blade washing. The time required for the research is 4 months (from April to July 2021), while the data for the analysis process is 15 days. Procedures to be followed when doing online blade washing: 1. Filling the blade washing tank, the steps are taken: a. Operate the make-up water transfer pump, b. Open the water supply valve 30SDD01AA101 to fill the blade washing tank, c. After the blade washing tank is filled to 1000mm, close the water supply valve 30SDD01AA101 (Blade washing pump auto stop level: 260 mm). 2. Gas turbine load setting at 200MW ±5MW. a. Make sure the operation mode select is in the “LOAD LIMIT” position. Done to anticipate unstable network frequency. b. Select APR MODE "OFF". JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 211 c. ALR SET at 200MW. d. Wait up to 30 minutes before performing online blade washing to stabilize the blade path temperature. 3. On-line washing a. Open valve 30SDD01AA102 (GT compressor blade washing pump suction Valve). b. Close valve 30SDD01AA106 (GT compressor blade washing pump min. flow orifice bypass valve). c. Close valve 30SDD01AA903 (GT compressor blade wash pump disc. line drain valve). d. Close valve 3*SDD01AA122 (GT comp blade wash water off-line supply a/b valve). e. Select the GT compressor blade washing pump “ON” push button in the local control panel. Make sure the pump is running and the “RUN” light is on. f. Make sure that “BLADE WASH AVAIL” is on the OPS. g. Select “ON-LINE WASH START” PB in OPS. h. Gently open and adjust the GT compressor blade washing pump discharge valve (3*SDD01AA103) to Pressure 5.4 kg/cm2 (or 0.15 m3/min). Keep the valve open and pay attention to the Pressure Indicator because if there is too much water flow it will cause the unit to trip. i. Make sure that the GT online washing water supply valve (3*SDD01AA702) is locally open. j. Water wash time for 3 minutes, pay attention to the blade path temperature. k. Select “OFF” PB in OPS. l. Make sure that the GT online washing water supply valve (3*SDD01AA702) is closed locally. m. Close the valve GT compressor blade washing pump discharge valve (3*SDD01AA103). n. Perform the steps ( g m ) up to 3 (three) times, with a pause of 10 minutes. o. Stop GT compressor blade washing pump when finished.. p. Hold gas turbine load at 200MW for 30 minutes. 4. Restoration a. After the GT compressor blade washing pump stops, open the GT compressor blade washing drain valves (30SDD01AA901, 902). Do not let any water remain in the pipeline for a long time. b. Close the GT compressor blade washing drain valves (30SDD01AA901, 902) after the remaining water is used up. c. Close valve 30SDD01AA102 (GT compressor blade washing pump suction valve). d. Open valve 30SDD01AA106 (GT compressor blade washing pump min. flow orifice bypass valve). e. Open valve 30SDD01AA903 (GT compressor blade wash pump disch. line drain valve). f. Open valve 3*SDD01AA122 (GT comp blade wash water off-line supply a/b valve). A literature study is used to obtain steps or formulas for the calculation process. Calculations were carried out according to the data obtained during observations and using formulas obtained from the literature. The results of this calculation will then be presented in the form of tables and graphs. This discussion contains the analysis of the results that have been obtained from the calculations that have been carried out. The conclusion is a final summary containing the results of the analysis, under with the research objectives. Gas turbine technical data and specifications: Manufacturer : X Model : M701F, single shaft JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 212 Rate output : 270 MW Efficiency : 38,2 % Type : axial flow type Fuel : Natural gas No. of stages : 4 Turbine inlet temperature : 1400°C Operating air temperature : 21,6°C – 35,5°C Max Loading rate : 6,7%/min Speed increase rate : 135 rpm/min Combustion chamber : 20 pcs, multi-can annular type Generator : 315 MVA Frequency : 50 Hz Power factor : 0,85 Speed : 3000 rpm Compressor Type : Axial flow type No. of stages : 17 Air flow : 651 kg/s Inlet air filter type : Static pressure Data retrieval is taken based on the results of several graphic forms that are stored in the OPS (operator station) memory continuously while the unit is operating. 3. RESULT AND DISCUSSION Table 1 describes the composition of the gas used for power generation. It is known that the fuel supplied from offshore PHE is natural gas with its composition and tabulated. Table 1. Consumption of gas fuel (PHE) Description Xi (mixture) mol (%) Mi (molal mass) kg/kmol BMf kg/kmol Carbon Dioxide, CO2 5.00 44.01 2.2005 Nitrogen, N2 0.61 28.02 0.170922 Methane, CH4 84.50 16.043 13.556335 Ethane, C2H6 4.91 30.07 1.476437 Propane, C3H8 2.88 44.097 1.2699936 ISOButane, iC4 0.79 58.124 0.4591796 N-Butane, nC4 0.60 58.124 0.348744 ISOPentane, iC5 0.27 72.151 0.1948077 N-Pentane, nC5 0.17 72.151 0.1226567 Hexane, C6 0.15 86.178 0.129267 heptane plus, C7+ 0.12 100.2 0.12024 Total 20.0490826 Lower calorific value (Low Heating Value) of fuel. 𝐶𝑎𝑙𝑜𝑟𝑖𝑓𝑖𝑐 𝑣𝑎𝑙𝑢𝑒 (𝐿𝐻𝑉) = 1115,6619 𝐵𝑇𝑈 𝑆𝐶𝐹 𝑏𝑒𝑐𝑎𝑢𝑠𝑒 ∶ 1 𝐵𝑇𝑈 = 1,0551 𝐾𝐽, 𝑎𝑛𝑑 1 𝑚3 = 35,315 𝑓𝑡 3 𝐶𝑎𝑙𝑜𝑟𝑖𝑓𝑖𝑐 𝑣𝑎𝑙𝑢𝑒 (𝐿𝐻𝑉) = 1115,6619 × 1,0551 × 35,315 = 41570,51796 ( 𝑘𝐽 𝑚3 ) The calorific value of fuel at the condition of entering the combustion chamber per unit volume. JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 213 𝐿𝐻𝑉𝑉 = 𝐿𝐻𝑉 ( 𝑃2×𝑇1 𝑃1×𝑇𝑓 ) (1) = 41570.51796 𝑘𝐽 𝑚3 ( 14,4 × 303,6 1,0286 × 473,2 ) = 373384.8790 𝑘𝐽 𝑚3 Gravimetric combustion value (LHVm). 𝐿𝐻𝑉𝑚 = 𝐿𝐻𝑉𝑉 × 𝜈 (2) = 𝐿𝐻𝑉𝑉 ( 𝑅𝑜×𝑇𝑓 𝐵𝑀𝑓×𝑃2 ) (3) = 373384.8790 𝑘𝐽 𝑚3 ( 1,16444 𝐽/𝑘𝑚𝑜𝑙 × 473,2 ˚𝐾 20,049 𝑘𝑔/𝑘𝑚𝑜𝑙 × 1412640 𝑁/𝑚2 ) = 51866.53878 𝑘𝐽/𝑘𝑔 Energy enters the system (Qin) Qin = ṁ𝑓 × 𝐿𝐻𝑉𝑚 (4) ṁ𝑓 = 65,613 𝐵𝐵𝑇𝑈 = 2733,859949 𝑀𝑀𝐵𝑇𝑈𝐻 1𝑀𝑀𝐵𝑇𝑈𝐻 = 27,49 𝑚3 ℎ , 𝜌𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑔𝑎𝑠 = 0,9 𝑘𝑔 𝑚3 (𝑡𝑎𝑏𝑙𝑒 𝐷𝑒𝑛𝑠𝑖𝑡𝑖𝑒𝑠 𝑜𝑓 𝐺𝑎𝑠) ṁ𝑓 = 2733,859949 × 27,49 × 0,9 [ 𝑘𝑔 𝑗𝑎𝑚 ] = 67638,42899 𝑘𝑔 𝑗𝑎𝑚 = 18,79 𝑘𝑔 𝑠 Qin = 18,79 𝑘𝑔 𝑠 × 2733,859949 𝑘𝐽 𝑘𝑔 𝑄𝑖𝑛 = 976220 𝑘𝐽/𝑠 = 976.22 MW By using the same method, the calculation results will be obtained as shown in Table 2. Table 2. Qin GT 3.1 NO T1 Mf LHV LHVv LHVm Qin K kg/s BTU/SCF kJ/m3 kJ/m3 kJ/kg MW 1 303,60 18,79 1.115,66 41.570,52 373.384,88 51.866,54 976,22 2 305,20 17,38 1.116,96 41.619,06 378.580,65 52.214,56 916,49 3 302,40 18,04 1.118,01 41.658,16 375.053,31 51.749,93 942,58 4 301,70 18,10 1.117,55 41.640,81 374.989,66 51.697,42 943,80 5 302,40 17,90 1.115,80 41.575,55 375.399,72 51.743,01 935,17 6 306,10 17,44 1.107,61 41.270,46 374.296,02 51.971,13 910,21 7 305,30 18,60 1.113,18 41.478,20 352.033,98 52.116,91 973,33 8 305,70 17,57 1.117,99 41.657,17 385.374,11 52.417,30 920,56 9 305,70 17,55 1.118,82 41.688,31 385.376,63 52.428,72 921,74 10 305,10 17,98 1.119,79 41.724,33 384.973,03 52.384,89 941,88 11 306,40 17,91 1.120,19 41.739,21 386.700,11 52.619,90 941,57 12 304,80 18,68 1.100,65 41.011,01 369.805,72 51.391,07 960,62 13 306,70 19,09 1.053,66 39.260,35 356.311,04 49.484,37 946,56 14 300,60 23,28 1.048,58 39.070,82 280.755,32 48.132,44 1.135,52 15 305,30 19,77 1.048,70 39.075,62 355.373,24 49.013,75 972,41 Stoichiometric mixed air requirements (100%) [A/F]th,m,d or ‘ air to fuel ratio ‘ theoretical gravimetric, dry is: | 𝐴 𝐹 | 𝑡ℎ,𝑚𝑜𝑙,𝑑 = Zc + 0,25. ZH + Zs – 0,5. Zo 0,21 𝑥 28,97 𝐵𝑀𝑓 JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 214 Calculated Zc, ZH, Zs, Zo dan ZN as follows: 𝑍𝑐 = 0.05(1) + 0.845(1) + 0.0491(2) + 0.0288(3) + 0.0079(4) + 0.006(4) + 0.0027(5) + 0.0017(5) + 0.0015(6) + 0.0012(7) 𝑍𝑐 = 1.1746 𝑚𝑜𝑙𝑒𝑠 𝑜𝑓 𝑐𝑎𝑟𝑏𝑜𝑛 𝑎𝑡𝑜𝑚𝑠 𝑝𝑒𝑟 𝑚𝑜𝑙𝑒 𝑍𝐻 = 0.845(4) + 0.0491(6) + 0.0288(8) + 0.0079(10) + 0.006(10) + 0.0027(12) + 0.0017(12) + 0.005(14) + 0.0012(16) = 4.137 𝑚𝑜𝑙𝑒𝑠 𝑜𝑓 ℎ𝑦𝑑𝑟𝑜𝑔𝑒𝑛 𝑎𝑡𝑜𝑚𝑠 𝑝𝑒𝑟 𝑚𝑜𝑙𝑒 𝑍𝑆 = 0 𝑍𝑂 = 0.005(2) = 0.1 𝑚𝑜𝑙𝑒𝑠 𝑜𝑓 𝑜𝑥𝑦𝑔𝑒𝑛 𝑎𝑡𝑜𝑚𝑠 𝑝𝑒𝑟 𝑚𝑜𝑙𝑒 𝑍𝑁 = 0.0061(2) = 0.0146 𝑚𝑜𝑙𝑒𝑠 𝑜𝑓 𝑛𝑖𝑡𝑟𝑜𝑔𝑒𝑛 𝑎𝑡𝑜𝑚𝑠 𝑝𝑒𝑟 𝑚𝑜𝑙𝑒 Air to fuel ratio theoretical, molar, dry or [A/F]th, mol, d: | 𝐴 𝐹 | 𝑡ℎ,𝑚𝑜𝑙,𝑑 = 1,1746 + 0,25 (4,137) + 0 – 0,5(0,1) 0,21 = 10,280 moles of air / moles of fuel Air to fuel ratio theoretical, mass, dry: | 𝐴 𝐹 | 𝑡ℎ,𝑚,𝑑 = | 𝐴 𝐹 | 𝑡ℎ,𝑚𝑜𝑙,𝑑 𝑥 28,97 𝐵𝑀𝑓 = 10,280 x (28,97/20,049) Kg of air/Kg of fuel = 14,854 kg.u/kg.bb A. Mass flow rate of air Figure 1 is the flow of air and fuel entering the combustion chamber in the gas turbine. Figure 1. Air, gas and fuel flow chart It is known from the manual book that the mass flow rate in an open cycle condition with 100% load is 2.188.300 kg/h = 607.861 (at T1 = 30°C). So for T1 = 30.6°C, m1 = 617.153 kg/s (calculated based on the ratio and the density of air). T1 : 30,6°C = 303,6 K m1 = ma : 617,153 kg/s mc : 47,24 kg/s T2 : 713,6K m2 = m1 – mc : 569,9132 kg/s TIT : 1400 ˚C (MHI manual book) mf : 18,7885 kg/s T3 : 1673,15K m3 = m2 + mf : mgp = 588,7016525 kg/s T4 : 864,4K m4 = m3 + mc : 635.941 kg/s JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 215 B. Actual ratio, dry air per mass of fuel Mass ratio, actual, dry air/fuel mass [A/F]act,m,d = m2 / mf = (569,9132 kg.u/s) / (18,7885 kg.bb/s) = 30,33316342 Kg air/kg fuel Molar ratio, actual, dry air/fuel mass: [ A F ] act, mol, d = [A/F]act,m,d 𝐵𝑀 𝑎𝑖𝑟/𝐵𝑀 𝑓𝑢𝑒𝑙 (5) = 30,33316342 28,97 20,049 = 20,9923 kmol air / kmol Mass ratio of fuel / air mass ƒ = 1 [ A F ] act, m, d = 1 30,33316342 = 0,032967218 𝑘𝑔. 𝑏𝑏 𝑘𝑔 . 𝑢 C. Percentage of excess air (excess air) Percentage of excess air = 100. (DC – 1), where DC = dilute coefficient DC = [A/F]actual [A/F]theoretical = 30,33316342 kgu/kgbb 14,854 𝑘𝑔𝑢/𝑘𝑔𝑏𝑏 = 2,042 Then % excess air (excess air) = 100 (2,042– 1) = 104,209%. So with (%) excess air of 104,209%, this means that the actual air requirement for the combustion process is 2,042 times the minimum theoretical air requirement, or 204% theoretical air is required. D. Compressor cycle calculation Actual compressor work per mass rate (Wkm) Figure 2. Compressor working process. 𝑊𝐾 𝑚 = (ℎ2 − ℎ1) (6) 𝑇1 = 30,6 ˚C = 303,6 K 𝑠𝑜 ∶ ℎ1 = 303,8116 kJ kg (𝑡𝑎𝑏𝑙𝑒 𝐴 − 17) 𝑇2 = 440,6 ˚C = 713,6 K 𝑠𝑜 ∶ ℎ2 = 727,9208 kJ kg (𝑡𝑎𝑏𝑙𝑒 𝐴 − 17) 𝑊𝐾 𝑚 = (727,9208 kJ kg − 303,8116 kJ kg ) 𝑊𝐾 𝑚 = 424,1092 kJ kg JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 216 Calculation 𝐂𝐏𝐚: specific heat of air at constant pressure kJ/kg.K At CPa = 0.997 − 1.022 kJ kg . K The commonly used CPa yang umum digunakan 1.005 kJ/kg.K = 101.325 kPa However, in this calculation, CPa uses 1.00926 kJ/kg.K Actual compressor power (PK) Pk = ma × CPa(T2 − T1) (7) = (617.1532 kg s × 1.00926 kJ KgK ) × (713.6 − 303.6)K = 255.37621 kJ/s = 255,37621 Megawatt E. Enthalpy of combustion in the combustion chamber For 204% theoretical air, the air temperature rise ΔT23 = 809 K. Because [A/F]m = (1/ƒ) = 30.33316342 kg.u/kg.bb or the actual combustion air requirement is 204% times the theoretical mixture, meaning [1+(1/30.33316342)] = 1,0329672 kg.gp/kg.u. The gas enthalpy from combustion is obtained from "Table A.3 Products theoretical air" with 200% and 400% theoretical air. The molar enthalpy at temperature T3 = 1673,15 K (200% and 400%): hˆ3,200% = 1889,022 kJ/kg (interpolasi) hˆ3,400% = 1846,468 kJ/kg (interpolasi) For the molar enthalpy T3 at 204%: hˆ3,204% = 1866,46838 kJ/kg (interpolasi) In the same way for T4 : hˆ4,204%= 916,144 kJ/kg Heat supplied Q204% = (1+ƒ) hiˆ3,204% - h2 = [(1,0329672) (1866,46838) – 727,9208] = 1235.452512 kJ/kg.u Specific heat of product gas (Cpgp) Is known: ΔT23 = 809°K Q204% = 1235,452512 KJ/Kg.u (1+ƒ) = 1,0329672 kg.gp/Kg.u h3 – h2 = [(1+ƒ) (Cpgp) (T3-T2)] = 945,079 kJ/Kg.u So the specific heat of the product gas Cpgp for theoretical air 204% is: Cpgp = 1235,452125 1,0329672 x 945,079 = 1,270655266 𝑘𝐽/𝑘𝑔𝑔𝑝 °𝐾 F. Turbine cycle calculation Actual work per mass rate (WT/m) Figure 3. Turbine work per mass rate JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 217 𝑊𝑇 𝑚 = (ℎ3 − ℎ4) (8) = hˆ3,204% − hˆ4,204% = (1866,648 − 916,144)𝑘𝐽/𝑘𝑔 = 1012,66𝑘𝐽/𝑘𝑔 Actual turbine power (PT) PT = mgp. Cpgp (T3 − T4) (9) = (588,7016525 kg s ) . (1,270655266 kJ kg . K) . (1673,15 − 864.4 K) = 604862,6006 𝑘𝐽/𝑠 = 604,8626006 𝑀𝑊 G. Calculation of gas turbine thermal efficiency (ɳth) ɳth = Pnet Qin = PT−PK Qin (10) = mg Cpgp(T3−T4)−ma Cpa(T2−T1) Qin (11) = 604,8626006 MW − 255,37621 MW 974,492 MW = 0,3586 𝑀𝑒𝑔𝑎𝑤𝑎𝑡𝑡 = 35,86% By using the same method, the calculation results will be obtained as shown in table 3, in table 3 this is taken for 15 days. Table 3. Thermal efficiency calculation results NO T1 PT Pk Pnet Qin ɳth °C MW MW MW MW % 1 30,6 604,8626 255.3762 350,510 976,2190 35,86 2 32,2 602,1951 268.7774 339,4774 916,4863 36,73 3 29,4 609,7126 267.2738 352,7663 942,5788 36,68 4 28,7 608,0405 265.7236 352,4559 943,8018 36,59 5 29,4 606,3278 266.8612 350,2702 935,1695 36,65 6 33,1 594,2965 264.9837 340,1773 910,2068 36,33 7 32,3 615,3465 262.9029 371,0327 973,3323 36,36 8 32,7 592,2283 259.7237 334,7549 920,5555 36,10 9 32,7 596,0175 263.8237 339,6523 921,7365 36,11 10 32,1 595,7878 257.2764 338,6659 941,8793 35,95 11 33,4 592,8049 255.7238 336,3591 941,5674 35,77 12 31,8 598,7885 254.2129 344,1901 960,6232 35,90 13 33,7 593,7196 254.6623 326,8501 946,5586 35,89 14 27,6 660,9847 254.9239 432,4037 1135,517 36,23 15 32,3 600,3166 253.2623 345,2760 972,4134 35,82 H. Effect of online blade washing on Pk (compressor work GT 3.1). Based on the data in Table 3, a graph of the relationship between compressor work and time (day) can be obtained as shown in Figure 4. JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 218 Figure 4. Compressor working graph AND Time (day) From the calculation data shown in Figure 5, it shows that there is an increase in compressor work after on-line blade washing (points 1-2). And there is a tendency for compressor work to decrease in the following days. I. The effect of online blade washing on fuel consumption The data in table 3 can also produce a graph of the relationship between fuel consumption and time (day) as shown in Figure 5. Figure 5. Graph of fuel consumption and time (day) In the trendline of Figure 5, it can be seen that after blade washing there is a fuel consumption savings of around 1.4 kg/s (points 1-2), then an increasing trend of fuel consumption can be seen in the following days. J. Effect of online blade washing on thermal efficiency (ƞTh) The relationship between compressor intake air temperature and thermal efficiency can be seen in Figure 6 (based on table data 3). 255,3762 268,77740 267,2738 265,7236 266,8612 264,9837 262,9029 259,7237 263,8237 257,2764 255,7238 254,2129 254,6623 254,9239 253,2623 250,0000 252,0000 254,0000 256,0000 258,0000 260,0000 262,0000 264,0000 266,0000 268,0000 270,0000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 C o m p re s s o r w o rk Time 18,78 17,38 18,04 18,09 17,9 17,44 18,6 17,57 17,55 17,98 17,91 18,68 19,09 18,67 19,76 y = 0,0239x2 - 0,3034x + 18,678 15 15,5 16 16,5 17 17,5 18 18,5 19 19,5 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 U s e g e B B (k g /s ) Time (day) JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 219 Figure 6. Graph of thermal and time (day) GT efficiency Figure 6 shows the change in thermal efficiency with time, which is shown in the black line. While the yellow line shows the trend line which clarifies the increase in thermal efficiency after blade washing (points 1-2) on the graph is ±0.8%. From the graph in Figure 6 online blade washing can maintain optimal GT performance, and will reduce the steep decline in GT performance when done regularly 4. CONCLUSION After calculating and analyzing, it can be concluded that 1) Cleaning the compressor blades using the online blade washing method can improve compressor performance on gas turbines, as indicated by the GT output power reaching 268,77738 MW and fuel savings of 1.4 kg/s. 2) Regularly doing online blade washing can improve GT performance with running hours until the next B Inspection (8000 hours) with an efficiency of 0.8%. In implementing online blade washing, it is made in the 52 Weekly plan so that the PIC is monitored and clear. REFERENCES 1. Setiawan B, Hidayat G, Cahyono SD. Analisis Pengaruh Compressor Washing Terhadap Efisiensi Kompresor Dan Efisiensi Thermal Turbin Gas. J Mesin Teknol (SINTEK Jurnal). 2017;11(1):49–54. 2. Sahlan, Irvan Buchari S. Taman RFA. Kegagalan Fungsi Safety Valve Lp Steam Drum HRSG 1.1 Muara Karang, doi.org/10.33322/powerplant.v3i1.809. J Powerpl [Internet]. 2015;3(1):48–55. Available from: https://stt-pln.e- journal.id/powerplant/article/view/809 3. Haouam A, Derbal C, Mzad H. Thermal performance of a gas turbine based on an exergy analysis, DOI:10.1051/e3sconf/201912801027. In: E3S Web of Conferences. 2019. 4. Krickis O, Zeltins N. Gas Combustion Efficiency Enhancement: Application Study of Intense Elestrostatic Field, DOI:10.2478/lpts-2019-0021. Latv J Phys Tech Sci. 2019;56(4):3–16. 5. Borello D, D’Angeli L, Salvagni A, Venturini P, Rispoli F. Study of particles deposition in gas turbine blades in presence of film cooling,DOI: 10.1115/GT2014-26250. Proc ASME Turbo Expo. 2014;5B(June). 6. Syofi’i I, Adanta D, Prakoso AP, Sari DP. Application of Computational Fluid Dynamics Method for Cross-flow Turbine in Pico Scale, DOI: https://doi.org/10.22219/jemmme.v6i1.12813. J Energy Mech Mater Manuf Eng. 35,86 36,73 36,68 36,59 36,65 36,33 36,36 36,1 36,11 35,95 35,77 35,9 35,89 36,23 35,82 y = -0,0022x2 - 0,0134x + 36,488 35,2 35,4 35,6 35,8 36 36,2 36,4 36,6 36,8 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 (E ff is ie n c y G T 3 .1 ( % ) Time (day) JEMMME (Journal of Energy, Mechanical, Material, and Manufacturing Engineering) Vol.6, No. 3, 2021 doi: 10.22219/jemmme.v6i3.18140 Wilarso | Online Blade Washing Analysis on Gas Turbine Performance in … 220 2021;6(1):1–8. 7. Rani S. Common failures in gas turbine blade: a critical review, DOI: 10.5281/zenodo.1207072. Int J Eng Sci Res Technol. 2018;7(3):799–803. 8. Bhagi LK, Gupta P, Rastogi V. A Brief Review on Failure of Turbine Blades, DOI: 10.13140/RG.2.1.4351.3768. Proc STME-2013 Smart Technol Mech Eng. 2013;(October):1–8. 9. Boyce MP, Gonzalez F. A Study of On-Line and Off-Line Turbine Washing to Optimize the Operation of a Gas Turbine, DOI:10.1115/1.2181180. J Eng Gas Turbines Power 129(1) [Internet]. 2007; Available from: https://www.researchgate.net/publication/245353326_A_Study_of_On- Line_and_Off- Line_Turbine_Washing_to_Optimize_the_Operation_of_a_Gas_Turbine 10. Mund FC, Pilidis P. Online compressor washing: A numerical survey of influencing parameters, DOI:10.1243/095765005X6881. In Proceedings of the Institution of Mechanical Engineers Part A Journal of Power and Energy 219(1):13-23; 2005. 11. Casari N, Pinelli M, Suman A, Vulpio A, Appleby C, Kyte S. Assessment of the Washing Effectiveness of on-Purpose Designed Eco-Friendly Cleaner Against Soot Deposits, DOI: 10.33737/jgpps/130789. J Glob Power Propuls Soc. 2020;4:253–63. 12. Brun KTGWCFRK. Experimental Evaluation of the Effectiveness of Online Water- Washing in Gas Turbine Compressors, DOI:10.1115/1.4028618. 2015; 13. Igie U, Pilidis P, Fouflias D, Ramsden K, Laskaridis P. Industrial Gas Turbine Performance: Compressor Fouling and On-Line Washing, DOI:10.1115/1.4027747. J Turbomach. 2014;136(10). 14. Mund FC, Pilidis P. A Review of Gas Turbine Online Washing Systems, DOI: DOI:10.1115/GT2004-53224. In Conference: ASME Turbo Expo 2004: Power for Land, Sea, and Air; 2004. 15. Rachmawatie D, Rustiadi E, Fauzi A, Juanda B. Empowerment of rural community through the development of renewable electricity, DOI: 10.22437/ppd.v8i2.8077. J Perspekt Pembiayaan dan Pembang Drh. 2020;8(2):139–54.