Microsoft Word - 66iervolino.docx


 CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 65, 2018 

A publication of 

 
The Italian Association 

of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors: Eliseo Ranzi, Mario Costa 
Copyright © 2018, AIDIC Servizi S.r.l. 
ISBN 978-88-95608-62-4; ISSN 2283-9216 

Design of a Fast Energy Storage and Energy Conversion 
System for Electric Vehicle 

Peng Wang 
School of Electrical Engineering, Xuchang University, Xuchang 461000, China 
Wangpeng461@163.com 

In this paper, the author researches on a fast energy storage and energy conversion system for electric 
vehicle. This paper analyses and studies several main energy storage technologies and expounds the energy 
storage principle, advantages and disadvantages which also presents the application status quo. This paper 
introduces the current situation of China's power system and the concept and connotation of smart grid in the 
future. Based on the background of the development and construction of smart grid in China, the application 
and development direction of energy storage technology are studied. In this paper, the structure of the battery 
energy storage system is introduced, and the mathematical model of the battery energy storage system is 
given. The power control strategy of converter is researched and designed. Combined with the development 
of wind power in China, the application and effect of battery energy storage system in wind power generation 
in China are studied and analysed. 

1. Introduction 

It is well known that the design of the ideal DC-DC converter involves many trade-offs. The increase in power 
density usually means the increase of overall power consumption, as well as the increase of the temperature 
of the junction temperature, the shell temperature and the PCB. In the same way, the optimization of DC/DC 
power for medium current to peak current almost always means sacrificing light load efficiency and vice versa 
(Yang et al., 2015; Errami et al., 2015). Because of the voltage difference between the instruments, the signal 
in the interconnected wire will add the pressure difference to the signal, causing a voltage "communication" in 
the wire. This is one of the reasons that audio signals hear 60 Hz noise (or horizontal interference in video 
signals). Another problem is the flow of current in the signal cable ground wire. The current will also be 
introduced into the cable and equipment. Therefore, it is the most basic requirement to ensure that the 
grounding loop current will not cause the problem of the system in the proper design of the grounding line in 
the system (Sun and Zhang, 2015). 
In another example, the grounding loop is a common problem when multiple audio - visual system 
components are connected together. The common noise in the audio system is often caused by the grounding 
loop problem. In addition, the "communication sound" can also be a typical grounding loop problem (of course, 
depending on the AC power voltage frequency used by the country in the country). Of course, the most 
common example of the ground loop problem is that the system uses instruments connected to the socket, 
while the other instrument connects different other grounding receptacles in the room. 
The DC/DC converter is a voltage converter that effectively outputs the fixed voltage after the input voltage is 
changed. The DC/DC converter is divided into three types: the boost type DC/DC converter, the step-down 
DC/DC converter and the up and down DC/DC converter. Three types of control can be used according to the 
demand. PWM has high control efficiency and has good output voltage ripple and noise. The PFM control 
model has the advantages of small power consumption even when it is used for a long time, especially in 
small load. PFM control is implemented in PWM/PFM conversion small load, and it is automatically converted 
to PWM control when heavy load is loaded. At present, DC-DC converter is widely used in mobile phone, 
MP3, digital camera, portable media player and other products. In the classification of circuit types, it belongs 
to the chopper circuit. Figure 1 show the basic circuits of DC/DC converter (Baloch et al., 2016; Tanvir et al., 
2015). 

463

                               
 
 

 

 
   

                                                  
DOI: 10.3303/CET1865078

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Peng Wang , 2018, Design of a fast energy storage and energy conversion system for electric vehicle, Chemical 
Engineering Transactions, 65, 463-468  DOI: 10.3303/CET1865078 



 

Figure 1: The basic circuits of DC/DC converter 

2. The structure of bidirectional power flow circuit for dc/dc converter 

The so-called Buck, Boost, Buck Boost and Cuk two-way power conversion circuit, that is, electric energy can 
flow from the input end to the output terminal, or from the output terminal to the input side, or the input port 
and the output terminal can be interchanged. To make the PWM DC/DC converter have this function, we must 
solve the problem of double current flow, and at the same time, when the current is in positive or negative 
direction, it will not change the equivalent circuit. How to make Buck, Boost, Buck ~ Boost and Cuk PWM 
DC/DC converter circuit has the function of it, the most simple way is in the original circuit (tube triode switch) 
on a parallel diode, a transistor in anti-parallel diode on the original circuit (switch), group a two can reverse 
conducting triode switch S and S. S and S work in a complementary way, that is, when S is turned on, the 
forward current flows through the original triode and the original diode. When S is turned on, the reverse 
current can flow through the transistor and diode. This will not only achieve positive and reverse current flow, 
do not make the equivalent circuit change, Buck, Bcost, Buck will check Boost and Cuk PWM to DC/DC 
converter circuit as shown in Figure 2, Boost, Buck Boost bidirectional Buck and Cuk PWM DC/DC converter 
circuit after transforming (Cárdenas et al., 2016). 

 

Figure 2: Cuk PWM to DC/DC converter circuit 

Reverse bidirectional Buck converter circuit and Boost converter M conversion ratio; reverse bidirectional 
Boost converter circuit and Buck converter with the same conversion ratio; bidirectional Buck ~ Boost 
converter and Cuk converter circuit, reverse conversion ratio is the same. 
The bidirectional converter circuit is used as the PWM DC/DC converter, and it has good characteristics in 
practical application. For example, the Cuk bidirectional converter can be used as a current charging 
converter, when DU=0.5, M=1, M=1, if Ui>U. At the time, the electric energy is Ui to U. Flow; when Ui, <U. At 
the time, the electric energy is U. Flow to Ui to keep Ui and U0. Always equal. 
Figure 3 shows the power bidirectional Cuk converter circuit and Cuk converter circuit is different: V1 switch 
(NPN type triode) at both ends of an anti-parallel diode diode DZ; D1 anti parallel a switch (PNP type triode 
V2). The base of the two switches is connected in parallel, and then the PWM pulse width modulation signal is 
connected with the current limiting resistance R. The circuit shown in Figure 3, if the right side is a positive and 
negative voltage source, and the left side is a load resistor, the electric energy will transmit to the left side from 
the right side, and the output voltage will be positive and negative. On the other hand, if the voltage source is 
at the left end (up and down negative) and the load resistance is connected to the right end, the electric 

464



energy is transferred from the left end to the right side, and the negative upper voltage is lower positive and 
negative. The circuit shown in figure 3 will appear a discontinuous conduction mode when the load current is 
12 hours. Diode will not flow backward through the current when the diode current drops to zero. But the 
circuit has a bypass tri - transistor that can lead the reverse current. Therefore, there is no discontinuous 
working mode in any small load current. This is very beneficial to the stable operation of the circuit. The so-
called complementary way of work is in the period of DUTS (DU is duty cycle, Ts is the switching period), the 
switch S opens, the S turns off; during (1 DU) Ts, the switch s opens, and the S turns off. 

 

Figure 3: Cuk PWM to DC/DC converter circuit 

3. System and algorithm 

In order to output the correct voltage value and manage power between different power sources and energy 
storage units, an efficient DC/DC converter must be used. The DC - DC converter converts a set of voltages 
and currents to different voltages and currents, and is limited by the maximum power. For example, 12 V turns 
400 V (and vice versa), and the maximum power in this case is 1.5 Kw. In this configuration, the battery is 
maintained for the dashboard actuator and lighting system. The standard voltage is 12V 200-800V, and the 
DC bus voltage is used to supply the power system. 
The technical specifications of bidirectional DC to DC converter are as follows: low voltage terminal nominal 
input voltage is 12V, during charging and discharging voltage, it may change from 8V to 16V. The nominal 
input voltage of high voltage terminal is 288V, and the working voltage range varies between 255V-425V. The 
nominal charge and discharge power is 1.5kW, and the switching frequency 50kHz is for safety. Electrical 
isolation between high and low voltage terminals is necessary. Then, the low voltage terminal of high-
frequency transformer is referred to as the earth. With respect to the low pressure end, the high end ground 
state is suspended. The system is composed of two full bridge connected by high-frequency transformer. In 
boost mode, the M1-M4 signal is controlled by PWM signal with duty ratio greater than 0.5, which is phase 
shifted compared with M2-M3 controller. 180 the leakage inductance of the transformer usually produces high 
voltage pulses on the main power devices, and the circuit loads of MCL DCL and CCL are responsible for the 
suppression of these pulses. The electric power stored in the clamping process of the clamp capacitor CCL is 
introduced into the main power circuit, which saves a large amount of electric energy than the traditional 
consumable buffer. In the depressurization mode, the converter uses a phase-shift modulation method to 
control the high pressure bridge arm by this way. According to transformer leakage inductance and output 
capacitance of the device, zero voltage switching operation (ZVS) is implemented in load range. Two direction 
PWM signals must be properly controlled, so that the converter can achieve maximum energy efficiency. To 
achieve the best current charging curve and improve the energy efficiency of the battery, there are three 
current sensors and two voltage sensors in the DC to DC converter control structure. Figure 4 shows the 
control system of the circuits.The add-edge-rule is designed base on the local consistency score criterion in 
the new algorithm. And the rule is embedded in the framework of ant colony algorithm. Therefore, the new 
algorithm can use the heuristic to dynamically reduce the search space and reduce the running time during 
the search process. The empirical tests show that without loss of results accuracy, the convergence speed of 
the proposed algorithm is significantly 40% faster than that of ant colony optimization. In addition, the 
Bayesian classifier based on constrained ant colony algorithm is proposed for project’s risk prediction of real 
estate project in the stage of decision. The basic equation is shown below: 

 
(1) 

,
( ) ( / ) ( , )p i i

x y
E f p x p y x f x y= 

465



̅( ) represents experience marginal distribution of x in the training sample. The expectation value calculated 
by model should be consistent with experience expectation value. 

 (2) 

C represents a series of probability distribution. The core idea of maximum entropy is to choose the model 
with largest entropy in these models. In all probability distribution, p* is selected, which meets the following 
equation. 

  
(3) 

H(p) represents condition entropy used to represent evenness of condition probability p(y|x). This is an 
optimization problem; the introduction of Lagrange operator makes us get the form of solution. 

  
(4) 

In the above formula 

 

(5) 

Take the formula (2) into formula (3) 

 

(6) 

By Bessel function, we have: 

 
(7) 

Based on the gradient descent method, node center and base width parameter are: 

   

(8) 

Two voltage sensors are used to adjust the voltage of the two battery, and two current sensor for current 
control of two battery charging curve, third current sensors are used to control the current transformer 
saturation, prevent power devices damaged in the explanation of these two control strategies, we should 
distinguish the charging process of the following: (1) high pressure the battery is responsible for maintenance 
of the battery charging; (2) the maintenance of the battery to charge the battery for high pressure. In the 
maintenance process of charging the battery, DC - DC converter in Buck mode, the voltage from 400 V to 12 
V, the low end switch does not work, only to achieve a continued flow diode rectifier voltage level of its built-in, 
four switch and PWM signal drive high voltage side of the selected control strategy for phase modulation 
method. Two fixed duty cycle of 50% complementary signals driving two switch on each arm, and between the 
two bridge arm signal phase angle by the control network determines the control strategy to prevent the use of 
symmetrical transformer core saturation adjustment, and buck converter duty cycle control, maintenance of 
battery charging by setting the dead time suitable for driving two of full bridge inverter complementary devices, 
the MOSFET is exactly zero voltage turn-on, so as to eliminate the conduction loss problem. As shown in 

{ / , {1, 2,..., }}p i ipC p E f E f i K= = ∈

,
( ) ( ) ( / ) log ( / )

x y
H p p x p y x p y x= − arg max ( )

p C
p H p∗

∈
=

1

1
( / ) exp ( , )

( )

K

i i
i

p y x f x y
Z x

λ∗
=

 
=  

 


1
( ) exp ( , )

K

i i
x i

Z x f x yλ
=

 
=  

 
 

2

( )
1

2
(2 )

( , )

mn mn

j mX nY
P

A jB

u X Y e dXdY
π π

π π

π
+

− −

+ =

 

1

1

1

2

2 sin ( )

0 2 sin

( ) sin

0

sin

0

6

1
[

6
1

]

mn mn

k M Y j mX nY

k M Y

jm n jmM Y jnY

jmM Y jnY

E
A jB

e dXdY

E
e e e dY

j m

e e dY

π π π α π

π π α π

πα π

π π

π

π π

π

+ − + +

+ − −

−

−

+ =

= −

 





( ) ( )
1

min 1, , 0
n

k
L d k k d n

=

 
= − + 

 


( ) ( ) ( ) ( )( )
( ) ( )( )

1

1 2
j j m j

j j

w k w k y k y k h

w k w k

η

α

= − + −

+ − − −
( ) ( )( )

2

3

j

j m j j
j

X C
b y k y k w h

b

 − 
Δ = −  

 
 

( ) ( )
( ) ( )( )

1

1 2
j j j

j j

b k b k b

b k b k

η

α

= − + Δ

+ − − −

466



Figure 5, when the M5 is turned from a pilot to a turn off, the M7 is still closed because of the dead time, and 
the center point of the half bridge will be suspended. 

 

Figure 4: The DC/DC converter control circuit 

 

Figure 5: PWM control 

Moreover, because the resonant circuit consisting of transformer leakage inductance and parasitic 
capacitance is located at the center point of the half bridge, it will generate natural oscillation, which will cause 
VDS6 to oscillate at fixed frequency. By setting the dead time correctly, M6 can be judged at zero voltage. 
Finally, in order to further improve the energy efficiency of the converter, M2-3 and M1-4 are driven by AND 
signals of the PWM signals of the high voltage terminals to achieve synchronous rectification function, so as to 
reduce the pressure drop when the freewheeling diodes are on the way. When charging the high pressure 
battery, the DC - DC converter sends electrical energy back to the high voltage battery group and increases 
the voltage from 12V to 400V. At this time, the high voltage end full bridge switch does not work, and the 
continuous current diode only performs the voltage rectifying function. At the low voltage end, the full bridge 
switch must be properly controlled to perform the boost operation and reduce the power loss to the maximum. 
The bootstrap circuit operation is implemented by a very convenient strategy: a two duty cycle PWM signal is 
higher than 50% to drive the two diagonal switch 180 degrees phase angle, this method allowing for an 
overlap period, all four switches are closed at the same time to charge the input inductor. At the same time, 
through the parallel two MOSFET bridge arms, the maximum resistance RD-Son is reduced to the maximum. 
In the remaining two cycles, another diagonal switch is directed, allowing the power to pass through the 
magnetic transformer by using the first and third quadrants. STM32F103 RC is the focus of the development 
of this application design, because the micro controller embedded with a powerful set of motor control 
peripherals, to allow very low CPU load to the system (double H bridge) to provide the correct vector of the 
core application design is the use of cross triggering unit (CTU) and eDMA (enhanced direct memory access), 
allowing users to trigger all the ADC capture and PWM configuration, without any CPU intervention in this 
case, a high degree of integration of motor control peripherals group (CTU-FlexPWM-ADC-DMA) that CPU 
performs control algorithm based on plant variables, namely the implementation of the control algorithm in 
each control period of 20 microseconds. Will CPU the actual occupancy rate decreased below 35%, such as 
the release of more performance to perform other tasks, and system security measures more (i.e. PWM 
generated /IO Redundant control) CTU activity is synchronized with the PWM cycle of the application, based 
on a pre-programmed list of instructions (two ADC units) that trigger automatic ADC acquisition.  
ADC CTU ADC are all collected in FIFO, once the value reaches a preprogrammed threshold, immediately 
trigger to transfer operation from SRAM DMA once and then the data pretreated by CPU (because the 

467



induction variable sampling completed) sent to the control algorithm, PWM control algorithm for vector control 
algorithm for fixed point the next control cycle, each direction conversion by two PID controller. Two ADC 
modules are implemented on the device. Each module contains user configurable sampling rate and 
conversion times. In this application, the total time of sampling and collecting data from the channel is less 
than 0.9 microseconds. The divider generates the ADC clock signal through the clock of the ADC digital 
interface. The internal simulation dog of ADC will compare the value of the converted data with the user 
programming threshold. If the converted data exceeds the threshold, an interrupt signal will be generated. 
Each ADC is controlled by the CPU (CPU control mode) and the CTU (CTU control mode). CTU can use an 
ADC command to control ADC sampling, but the premise is that ADC is in the CTU control mode. In this case, 
two ADC acquisition operations can be triggered at the same time. In this application, two ADC collects three 
current values and two voltage values, and converts them to 10 bit resolution data (maximum voltage 3.3V). 
Figure 6 shows the result. 

 

Figure 6: The control result 

4. Conclusion 

The cycle of PWM signal involved in this application is 20 microseconds (DC to DC switching frequency 50 
kHz), so select sub module 0 to synchronize / reload all PWM sub modules, so as to avoid losing match 
between PWM signals. The same module is also used to restart the CTU control cycle. Phase modulation (HV 
H bridge) PWM signal is 0 and 1 sub modules, each module produces two signals, each signal (in the same 
sub module) are complementary (only different dead time). The sub module 1 has a delay between the 
generated signal of the sub module 0, and the delay time is a phase-shift modulation value generated by the 
DC/DC control algorithm. The synchronous rectifier is based on the two signals generated by the sub module 
2. Duty ratio modulation based on sub module 2 signal (like phase modulation synchronous rectifier one) 
sample. The signal duty ratio varies in the range of [52%, 90%], and the second signal is delayed by half a 
period (10 microseconds). The sub module 3 provides a dual switch output signal to generate the required 
active clamp PWM signal. 

References 

Baloch M.H., Wang J., Kaloi G.S., 2016, Stability and nonlinear controller analysis of wind energy conversion 
system with random wind speed, International Journal of Electrical Power & Energy Systems, 79(11), 75-
83, DOI: 10.1016/j.ijepes.2016.01.018. 

Cárdenas R., Díaz M., Rojas F., Clare J., 2016, Resonant control system for low-voltage ride-through in wind 
energy conversion systems, IET Power Electronics, 9(6), 1297-1305, DOI: 10.1049/iet-pel.2015.0488. 

Errami Y., Ouassaid M., Maaroufi M., 2015, A performance comparison of a nonlinear and a linear control for 
grid connected PMSG wind energy conversion system. International Journal of Electrical Power & Energy 
Systems, 68, 180-194, DOI: 10.1016/j.ijepes.2014.12.027. 

Sun L., Zhang N., 2015, Design, implementation and characterization of a novel bi-directional energy 
conversion system on DC motor drive using super-capacitors, Applied Energy, 153, 101-111, DOI: 
10.1016/j.apenergy.2014.06.084. 

Tanvir A.A., Merabet A., Beguenane R., 2015, Real-Time Control of Active and Reactive Power for Doubly 
Fed Induction Generator (DFIG)-Based Wind Energy Conversion System, Energies, 2015(8), 10389-
10408, DOI: 10.3390/en80910389. 

Yang Y., Mok, K.T., Tan, S.C., Hui, S.Y., 2015, Nonlinear Dynamic Power Tracking of Low-Power Wind 
Energy Conversion System, IEEE Transactions on Power Electronics, 2015, 30(9), 5223-5236, DOI: 
10.1109/TPEL.2014.2363561. 

468