Design of a new supercapacitor energy management system for electric vehicles

The energy management system composed of supercapacitors and batteries takes into account the advantages of high power density of supercapacitors and high energy density of batteries, which can better meet the requirements of electric vehicle starting and acceleration performance, improve the recovery efficiency of electric vehicle braking energy, and increase driving range.

1 System Overview
The supercapacitor and battery energy management system mainly consists of two parts: BLDCM drive controller and bidirectional DC-DC circuit. The system block diagram is shown in Figure 1.

In Figure 1, L, M1, and M2 form a bidirectional DC-DC circuit, VT1 to VT6 form a three-phase inverter, and a high-end load switch M3 is used to control the on and off of the bus and battery when necessary. The battery bus voltage Vin = 72 V, the supercapacitor rated parameters are 165 F/48 V, and the brushless DC motor parameters are 72 V/5.5 kW. When the motor is running, the load switch M3 is turned on, the three-phase inverter works normally, the bidirectional DC-DC does not work, and the system energy comes from the battery; when the motor energy is fed back for braking, the bus voltage is higher than the battery voltage, and the comparator C1 signal triggers the shutdown of the load switch M3, the bidirectional DC-DC works in the BUCK state, and the supercapacitor is charged; when the motor is started or a large torque is output, the bidirectional DC-DC works in the BOOST state, which generally lasts only tens of seconds. When the supercapacitor has sufficient energy, it can ensure that the BOOST output voltage is higher than the bus voltage, and the load switch M3 is turned off. If the discharge time is too long, since the supercapacitor does not have a constant voltage characteristic, its terminal voltage will continue to decrease as energy is consumed, and the output voltage of the corresponding BOOST circuit will also decrease accordingly. When the output voltage value is smaller than the bus voltage value, the high-end load switch M3 is turned on. At this time, the battery alone powers the system and shuts down the bidirectional DC-DC circuit of the supercapacitor part.

2 System working principle and control strategy
2.1 Bidirectional DC-DC principle
The reasons why this system uses a bidirectional DC-DC converter are: (1) the voltage at the supercapacitor terminal does not match the battery voltage; (2) the supercapacitor does not have a constant voltage characteristic. Since the voltage characteristics are inconsistent with those of the battery, the two cannot be directly connected in parallel. The rated voltage of the supercapacitor used in the system is 48 V and the rated voltage of the battery is 72 V, so the low-end voltage of the bidirectional DC-DC converter is 48 V and the high-end voltage is 72 V. Since the voltage conversion range is not large, there is no need to use a transformer for voltage conversion, and PWM chopping can be used directly. The bidirectional DC-DC structure is shown in Figure 2.
The bidirectional DC-DC converter in Figure 2 is essentially a combination of the basic BUCK circuit and the BOOST circuit [1]. The power diode in the BUCK circuit or the BOOST circuit is replaced by a power MOSFET to obtain the circuit topology shown in Figure 3. Depending on the direction of energy flow, the circuit operates in the BUCK buck mode or the BOOST boost mode.
In the BUCK step-down mode, the M1 tube is used as a switch tube, and the driving signal comes from the PWM control chip; the M2 tube is used as a diode, and the parasitic body diode of the M2 tube is used. At this time, the M2 must be reliably turned off by negative voltage to achieve reliable operation of the circuit. The circuit is set to work in CCM mode, and the equivalent circuit in the step-down mode is shown in Figure 3. The arrows in Figure 3 represent the direction of voltage and current, and energy flows from V1 to V2, which is the charging mode of the supercapacitor. The time period t0~t1 indicates that M1 is turned on, and the time period t1~t2 indicates that M1 is turned off. Assuming the PWM period is T and the duty cycle is D, the M1 turn-on time is DT and the M1 turn-off time is (1-D)T. According to the inductor volt-second balance principle, the average value of the volt-second value at both ends of the inductor L in one cycle is 0, so the average volt-second value of the inductor in one cycle can be obtained by the following formula:

In the BOOST boost mode, the M2 tube is used as a switch tube, and the driving signal comes from the PWM control chip; the M1 tube is used as a diode, and the parasitic body diode of the M1 tube is used. At this time, M1 must be reliably turned off by negative voltage to achieve reliable operation of the circuit. The circuit is set to work in CCM mode, and the equivalent circuit in boost mode is shown in Figure 4. The arrows in the figure indicate the direction of voltage and current, and energy flows from V2 to V1, which is the discharge mode of the supercapacitor. The time period t0~t1 indicates that M2 is turned on, and the time period t1~t2 indicates that M2 is turned off. Assuming the PWM period is T and the duty cycle is D, the M2 turn-on time is DT and the M2 turn-off time is (1-D)T. According to the inductor volt-second balance principle, the average value of the volt-second value at both ends of the inductor L in one cycle is 0, so the average volt-second value of the inductor in one cycle can be obtained by the following formula:

Since the duty cycle is 0V2, that is, V2 is obtained by PWM chopping to obtain the bus voltage V1 that meets the motor working requirements.
2.2 Energy management system control strategy and working mode
2.2.1 Design requirements
The electric vehicle energy management system has high requirements for safety and should meet the following conditions:
(1) Meet the safety requirements of braking and acceleration and conform to the driver's habits. By finding the optimal coverage range of electronic brakes and mechanical brakes, the energy can be recovered to the maximum extent while ensuring safety. The electric braking process with energy recovery system should be as similar as possible to the traditional braking process; during the acceleration process, as much energy as possible is released to ensure the required acceleration performance of the vehicle.
(2) Considering the performance of the energy management system and the motor, ensure the safety of components such as supercapacitors, inductors, and motors during energy feedback and release, and avoid damage to components due to excessive charging and discharging currents or excessive charging voltages.
2.2.2 Control strategy
(1) Energy feedback control strategy
Under the condition of meeting the design requirement (1), determine the optimal braking force according to the limit value of requirement (2) to maximize the recovered energy, that is, the integral of current over time is maximized. In order to conform to the usual braking habits, the brake pedal is reused for electric braking and mechanical braking. The entire brake pedal stroke is divided into two sections. The first section is the electric braking control section. As the pedal goes down, the electric braking strength gradually increases; the second section is the mechanical braking control section. As the pedal goes down, the mechanical braking strength gradually increases.
Each limiting factor is quantified as the current maximum allowable braking torque, and this is used to limit the braking torque of the motor, thereby protecting the normal operation of the system. The limiting factors of electric braking mainly come from two aspects: the motor and the energy management system, including the maximum allowable braking torque of the motor, the maximum allowable braking power of the motor, the maximum allowable charging power of the energy management system, and the maximum allowable charging current of the energy management system. The specific strategy for converting these limiting factors into motor torque limitation is:

In the formula, each physical quantity is a positive value; min() means taking the minimum value; max() means taking the maximum value, Pmmax means the maximum allowable braking power of the motor; Pbmax means the maximum allowable charging power of the energy management system; Ibmax means the maximum allowable charging current of the energy management system; Vb means the current terminal voltage of the energy management system. The two limiting factors and the terminal voltage of the energy management system are variables, and the current transient values ​​of the system operation are taken and given by the energy management system; the motor power generation efficiency and the current motor speed are variables, and the current transient values ​​of the motor operation are taken and given by the motor control system.
(2) Energy release control strategy
The specific description of the energy release control strategy is similar to the energy feedback control strategy. Each limiting factor is quantified as the current maximum allowable driving torque, and this is used to limit the driving torque of the motor, thereby ensuring the normal operation of the system.
3 Bidirectional DC-DC control method
The bidirectional DC-DC control method adopts voltage and current dual closed-loop control [2], in which the voltage loop is the outer loop, and the closed-loop control of the voltage is realized by TL431 and optocoupler; the current loop is the inner loop, and the closed-loop control method of the peak current is adopted. Peak current control not only has a fast response speed, but also has a current limiting protection function, which can improve the reliability of the system. The basic principle of peak current control is shown in Figure 5. Figure 5 (a) shows the peak current control principle in BUCK mode, and the peak current control principle in BOOST mode is similar to it. In the figure, the error signal obtained by subtracting the reference voltage Vref from the converter output voltage V(t) is amplified by the compensation network as the modulation signal of the PWM modulator, and the current sampling signal is(t)Rf is used as the carrier signal. At the beginning of each switching cycle, the RS trigger is set by the clock pulse, the switch device M1 is turned on, and then the inductor current gradually increases, as shown in Figure 5(b). When it is detected that the current signal is(t)Rf is greater than the modulation signal ic(t)Rf, the comparator reverses and resets the RS trigger, so that the power tube switch is turned off, and the inductor current is continued through the freewheeling tube. Figure 5(b) shows the PWM duty cycle change waveform under two inductance and current growth slope conditions. The waveform in the figure shows that when the inductance and current grow fast (large slope), that is, when the load is large (for supercapacitor charging, it is the initial charging moment, the circuit is close to a short-circuit state), the current quickly reaches the peak value, and the circuit quickly enters the peak current control state, which is manifested in the duty cycle of the PWM output waveform becoming smaller; conversely, the duty cycle of the PWM output waveform becomes larger.

4 Hardware Design of Bidirectional DC-DC
In this design, a dual closed-loop structure is used to control current and voltage. The control chip is UCC3803A from TI. An error amplifier and current amplifier inside UCC3803A can easily form a dual closed loop of current and voltage. In actual use, in order to have a faster response speed, the error amplifier can be omitted, and the voltage regulator TL431 and the optocoupler PC817 are used to form a voltage feedback. The current loop is formed by using the current sensor LAH 100-P from LEM. The BUCK control circuit is shown in Figure 6, and the principle of the BOOST control circuit is similar to it, except that the current direction and the position of the switch tube are changed. IS1 is the voltage measurement signal output from the LEM Hall current sensor LAH 100-P. The current signal enters the current feedback terminal, that is, the ISEN terminal in Figure 6. V48 comes from the output of the power part. Since the maximum voltage of TL431 can only be regulated to 36 V, it is necessary to partially modify the classic TL431 voltage regulator circuit to meet the 48 V voltage regulation requirements. Therefore, a 24 V voltage regulator tube is introduced at the 3rd foot (i.e., K pole) of TL431. The terminal voltage of TL431 is about 24 V, which can normally play a voltage regulation role within the safe working area. PC817 realizes electrical isolation and stabilizes the voltage through the output voltage Vce. When the supercapacitor voltage approaches 48 V, the PC817 output current Ic increases, and Vce decreases. The signal entering the 2nd foot VFB compensation terminal of UCC3803 will also decrease, and the PWM output duty cycle will also decrease accordingly; when the supercapacitor voltage exceeds 48 V, the 1st foot of the UCC3803 compensation terminal is pulled low, and the PWM is turned off, which plays the role of overvoltage protection. At this time, the circuit will maintain a dynamic balance at 48 V.

This system is currently undergoing experimental verification, and its operation is stable and its energy feedback and release performance are good.