This article is about 10,000 words, it is recommended to save it for reading

Author | Dou Mingjia

Produced by | Automotive Electronics and Software

On September 22, 2020, at the 75th United Nations General Assembly, the Chinese government proposed: "China will increase its national independent contribution and adopt more powerful policies and measures to strive to peak carbon dioxide emissions by 2030 and strive to achieve carbon neutrality before 2060." The realization of carbon neutrality must promote the simultaneous transformation of upstream high-energy-consuming industries and downstream applications, and hydrogen energy is an important way to achieve carbon neutrality. In the application of downstream transportation links, although the number of diesel trucks accounts for only 7.95 of the total number of vehicles, their emissions of carbon monoxide, carbon oxides, nitrogen oxides and particulate matter account for 10%, 18.8%, 60% and more than 84.6% of the total vehicle emissions. In order to achieve carbon neutrality, electrification in the transportation field is a general trend, and the electrification of commercial vehicles, especially diesel trucks, is an important accelerator towards carbon neutrality.

In the usage scenarios of new energy commercial vehicles, pure electric will become the main route for urban distribution logistics within 400km and short transportation distances, multiple start-stop conditions, and the implementation of new energy road rights policies. Pure electric will first penetrate into urban distribution light trucks, municipal sanitation, and short-distance traction. For inter-city long-distance transportation scenarios such as long-distance traction of more than 400km and with heavy loads, pure electric models are difficult to realize due to their high requirements for battery capacity and power consumption. In the long run, hydrogen fuel cell models will dominate.

In addition, when purchasing a vehicle, commercial vehicle users not only focus on the purchase cost of the vehicle, but also on the total cost of ownership (TCO), including procurement cost, operating cost, maintenance cost and management cost. Among them, operating cost accounts for the largest proportion of TCO, and fuel cost accounts for the largest proportion of operating cost. Therefore, for hydrogen fuel heavy trucks, while the country is promoting the reduction of hydrogen costs, reducing the hydrogen consumption of each vehicle will bring huge benefits to its TCO reduction.

The commonly used configuration of fuel cell vehicles is as above, which includes the following main parts:

1. Fuel cell system: Taking the proton exchange membrane fuel cell PEMFC (Proton Exchange Membrane Fuel Cell) as an example, it includes a fuel cell stack (a combination of multiple single cells), an air intake system (air filter, flow meter, air compressor, intercooler, etc.), a hydrogen system (hydrogen circulation pump/ejector, hydrogen injector, etc.), and a thermal management system (water pump, fan, water tank, etc.);

   
   

2. Fuel cell DCDC: The output voltage of the fuel cell stack (depending on the number of cells and the voltage of the cells) usually cannot meet the working voltage requirements of the high-voltage components of the vehicle, and a DCDC is required for boosting the output;

   
   

3. Hydrogen storage system: The main energy source of fuel cell vehicles comes from hydrogen (H 2 ) stored in on-board hydrogen storage bottles . Usually, different numbers of hydrogen bottles are equipped according to the vehicle's mileage requirements and the hydrogen storage capacity of a single hydrogen bottle (35Mpa/70Mpa);

   
   

4. High-voltage power battery: Due to the complex working conditions of the vehicle during driving, there will be frequent acceleration, climbing, braking, etc., and the fuel cell cannot meet the power demand of the whole vehicle in real time due to its own characteristics, so a high-voltage power battery is needed to "cut the peak and fill the valley";

   
   

5. Drive system: For heavy trucks, it can be a dual integrated axle or distributed (4 motors) drive system, which requires a motor controller, drive motor, gearbox, differential, etc. to convert electrical energy into mechanical energy to drive the vehicle.

   
   

6. Control system: including fuel cell control system, hydrogen system controller, high-voltage power battery control system, motor control system and vehicle controller, etc.;

In the above configuration, the fuel cell system and the high-voltage battery system serve as the power source of the vehicle, and they have different working characteristics:

2.1 Analysis of fuel cell system characteristics

1. Durability of fuel cells

The service life of fuel cells has always been a key problem that restricts the large-scale application of fuel cell vehicles. For fuel cell commercial vehicles, their on-board fuel cells should have a normal service life of at least 20,000 hours. However, this requirement is usually not met under vehicle operating conditions. The main degradation conditions that lead to the reduction of fuel cell life are:

• Frequent start-stop conditions: Frequent start-stop conditions will exacerbate the life of the fuel cell system. Therefore, the energy management strategy needs to reasonably control and reduce the frequency of fuel cell start-stop conditions to extend its service life.

• Continuous low-load or idling conditions: When the fuel cell is in a continuous low-load or idling condition, the single cell of the stack will maintain a high potential state close to the open circuit voltage. At this time, the cathode potential is usually 0.85V-0.9V. This high potential state will cause chemical corrosion of the proton exchange membrane, thereby causing irreversible life decline of the stack. Therefore, the energy management strategy needs to keep the fuel cell working in the high-efficiency area as much as possible under low load conditions, and reduce the occurrence of low-load or idling conditions by charging the high-voltage power battery to extend the service life of the fuel cell;

• Large load-changing conditions: Due to the large load capacity of heavy trucks, complex road conditions, and large load fluctuation range, when the fuel cell is in a large load-changing condition, there will be insufficient supply of reaction gas, which will cause life degradation. In addition, continuous load changes will further aggravate the damage of the catalyst and carbon carrier inside the fuel cell stack. Therefore, the energy management strategy needs to reduce the large load-changing conditions of the fuel cell system and slow down the degradation of the fuel cell life;

• Overload conditions: Fuel cells have a certain overload capacity, but long-term full-load operation will cause the fuel cell life to decrease. Therefore, the energy management strategy needs to avoid long-term high-load or overload operation of the fuel cell.

2. Fuel cell efficiency analysis

From the power-efficiency curve of the fuel cell below, it can be seen that when the output power of the fuel cell is smaller than point c, its efficiency is low. When the output power continues to decrease, its efficiency will drop sharply. When its output power is greater than point d, the system efficiency will also show an accelerated downward trend. Therefore, the energy management strategy should limit the working range of the fuel cell to the high efficiency zone (cd) and try to maintain it in the optimal efficiency zone (ab).

Fuel cell operating characteristic curve

Fuel cell power-efficiency curve

3.1 Optimization of power distribution between fuel cell and high voltage battery

Based on the above description of the basic configuration of hydrogen fuel heavy trucks and the working characteristics of the two power sources (fuel cell + high-voltage battery), how to reasonably allocate the output power of the fuel cell and the output power of the high-voltage battery under different operating conditions of hydrogen fuel heavy trucks, so as to meet the power requirements of the whole vehicle, increase the life of the fuel cell and minimize hydrogen consumption. There are different control strategies in the industry for this problem, which are mainly divided into the following categories:

• Switch control: simple and easy to implement. Its core is to always control the fuel cell to work in the high-efficiency range. When the power battery SOC reaches the set maximum value, the fuel cell is turned off and the power battery works alone. When the power battery SOC is lower than the set value, the fuel cell is turned on and controlled to charge the power battery while meeting the power demand of the vehicle, until the power battery SOC reaches the maximum value and then the fuel cell is turned off again, and the power battery works alone. This cycle is repeated continuously. With this control strategy, the power battery will continue to be charged and discharged at a deep level, and the fuel cell will start and stop frequently. This strategy will affect the life of the power battery and the fuel cell.

• Power following control: The core is to control the SOC of the power battery to approach the target expected value, while the fuel cell adjusts the output power in real time within the allowed range according to the SOC of the power battery and the required power of the vehicle;

• Fuzzy control: does not require an accurate mathematical model and has good adaptability to fuel cells with complex dynamic characteristics. However, this strategy relies more on the accumulation of engineering experience, such as the fuzzy control strategy based on WAFA.

• Instantaneous optimization - equivalent hydrogen consumption minimum control: Establish an instantaneous equivalent hydrogen consumption model for the entire vehicle and establish a corresponding real-time objective function, always taking instantaneous hydrogen consumption minimum as the control target, so as to distribute the power of the fuel cell and the power battery, such as equivalent consumption minimum ECMS (Equivalent Consumption Minimization Strategy);

• Global optimal control: Generally, it aims to achieve the control objectives of global minimum energy consumption under fixed typical operating conditions. The core lies in establishing an accurate mathematical model or prediction model to achieve optimal control under specific conditions through complex calculations. The control effect is greatly affected by the accuracy of the model and is difficult to achieve, such as Predictive Adaptive Cruise Control (PACC).

3.1.1 Power following control

The power following control is mainly based on the output power of the fuel cell, and the power battery supplements the peak power. According to the current SOC of the high-voltage power battery, it is mainly divided into three modes:

1. SOC＜SOC min

At this time, the SCO of the power battery is low, and the power output of the power battery should be avoided or reduced as much as possible. The power demand of the heavy truck is all provided by the fuel cell, and the power battery is charged until the SOC of the power battery>SOCmin . At this stage, the fuel cell outputs the maximum power Pfc = Pfc -max ;

2. SOC min ＜SOC ＜SOC max

At this time, the power battery is in the ideal working range and can be properly charged or discharged. The control method at this time is to make the fuel cell work in its high efficiency area as much as possible and meet the power requirements of the vehicle as much as possible. The power battery supplements the peak power or recovers the excess energy. At this time, the output power of the fuel cell varies within the range of:

P fc-min ≤P fc ≤P fc-max

3. SOC＞SOC max

At this time, the power battery should not continue to be charged, otherwise it will affect the effect of energy recovery. The control method at this time is: maintain the fuel cell in the high-efficiency area and control the power output as small as possible to speed up the power consumption of the power battery and make the power battery SOC return to a reasonable range as soon as possible.

3.1.2 Energy management strategy based on WAFA composite fuzzy control

Although the above-mentioned power following control energy management strategy has the advantages of simplicity and reliability, it cannot adapt to the dynamic working conditions of heavy-duty trucks. Fuzzy control is a nonlinear extension of logic threshold control and is a highly robust nonlinear control. It has good adaptability to fuel cell heavy-duty trucks with complex dynamic working conditions.

WAFA compound fuzzy control logic structure

The WAFA composite fuzzy control strategy is based on a single fuzzy control system to meet the power requirements of the heavy-duty truck power system. It also dynamically corrects the output power of the fuel cell and the power battery by adding a sub-fuzzy control system to achieve real-time dynamic adjustment of the power battery SOC. On this basis, the weighted mean filtering algorithm WAFA is used to smooth the fuel cell power output by the composite fuzzy control to reduce the output of high-frequency power, and the load rate limiting method is used to limit the load rate of the fuel cell to avoid the occurrence of a significant degradation condition. This strategy achieves a reasonable distribution of power between the two power sources while taking into account the economy and durability of the fuel cell and the steady-state regulation capability of the power battery.

1. Design of main fuzzy controller

The function of the main fuzzy controller is to determine the output power of the fuel cell according to the power demand of the power system and the SOC of the power battery, so that the power system can meet the requirements of heavy truck power and economy. First, it is necessary to determine the membership function of the three variables: input variable power system power demand Prep , power battery SOC and output variable fuel cell output power Pfc, that is, to set the basic domain.

Main fuzzy controller system interface

The domain of the membership function of the power system demand power Prep is set to [0,300], and the membership function is divided into five fuzzy subsets {VS, S, M, B, VB}, where VS, S, M, B, VB represent very small, small, medium, large, and very large, respectively.

The domain of the power battery SOC is set to [0,1], and the membership function of SOC is divided into five fuzzy subsets {XS, S, M, B, XL}, where XS, S, M, B, XL represent very small, small, medium, large, and very large, respectively.

The domain of the membership function of the fuel cell output power Pfc is defined as [20, 75]. Pfc is divided into five fuzzy subsets: {SS, S, M, B, BB}, where SS, S, M, B, BB mean very small, small, medium, large, and very large, respectively.

The design objectives of formulating fuzzy control rules are as follows:

• Reduce the number of starts and stops of the fuel cell;

• Ensure that the fuel cell operates in a high efficiency range, minimize the possibility of hydrogen depletion, and try to operate in the optimal efficiency range;

• Overdischarging of power batteries will reduce the cycle life of power batteries. The energy management strategy should limit the power battery SOC to a reasonable operating range and possibly avoid the power battery from being charged and discharged at a high rate.

• When a heavy truck is in high power demand such as starting, accelerating, climbing, etc., the power battery should provide instantaneous peak power to the vehicle in a timely manner;

• When the heavy truck is in deceleration and braking state, the power battery is used to recover the braking energy;

According to the above objectives, the IF-THEN rule table is as follows:

Main fuzzy controller IF-THEN rule table

The power distribution diagram of the fuzzy controller obtained by the above control rules is as follows:

Power distribution of the main fuzzy controller

2. Sub-fuzzy controller design

The function of the sub-fuzzy controller is to address the shortcomings of large SOC fluctuations and hysteresis in the heavy-duty truck power battery. According to the difference between the power battery SOC and the target SOC, the fuel cell power output by the main fuzzy controller is dynamically corrected to make the power system adjustment more flexible to adapt to the working conditions of heavy-duty trucks.

Sub-fuzzy controller system interface

The sub-fuzzy controller takes the difference ΔSOC between the current SOC of the power battery and its expected value SOC* as input, and the output power correction coefficient α of the fuel cell as output. According to the calculation and control requirements, the domain of ΔSOC is set to [-0.5, 0.5], and its membership function is divided into three fuzzy subsets {Small, Middle, Big}, representing low, medium, and high; the domain of α is set to [0, 2], and its membership function is divided into three fuzzy subsets {S, M, L}, representing small, medium, and large. The control rules of the sub-fuzzy controller are formulated to obtain the distribution diagram of the correction coefficient of the fuel cell output power.

Rule curve of correction coefficient α

3. Design of control algorithm for deterioration condition

Since the details of fuzzy control are uncontrollable, the fuel cell under this strategy is prone to degradation. In order to increase the service life of the fuel cell, a control method based on fuel cell protection priority is used to supplement the fuzzy control strategy according to the control requirements.

(1) Weighted mean filtering algorithm smoothes the output power of the fuel cell

Frequent fluctuations in the output power of a fuel cell will aggravate its life attenuation. Therefore, the output power is smoothed by a weighted mean filtering algorithm to reduce high-frequency power output. The power output signal of a fuel cell is a set of discrete time signals. The filter window continuously slides along the discrete time series to sample N data. Each time the sample is taken, the filter window is moved once, and then the power values ​​corresponding to the N sampling moments in the filter window are weighted averaged. The weight of each power value is determined by the time difference between its sampling moment and the current sampling moment t. The larger the difference, the smaller the weight.

Schematic diagram of weighted mean filtering algorithm

By taking a weighted average of the N values ​​in the filter window after each sliding, a set of smoothed power output signals can be obtained. Under the premise of meeting the transient response capability of the fuel cell heavy-duty truck power system, the N value is selected by comprehensively considering the fuel cell output power load rate range and the peak charge and discharge capacity of the power battery. If the value of N is too small, the purpose of smoothing the fuel cell output power will not be achieved. If the value is too large, the fuel cell output power will be over-smoothed, which will affect the normal output capacity of the fuel cell and reduce the performance of the whole vehicle. In order to take into account the smoothness of the fuel cell output power and the normal power output capacity, this paper selects the number of data N that the sliding filter window can accommodate as 4.

4. Limit the load rate of fuel cell output power

Large load changes will exacerbate the life degradation of the fuel cell. To protect the fuel cell, the rate of change of the smoothed output power is usually further limited based on the speed limit provided by the fuel cell supplier.

3.1.3 ECMS energy control strategy with minimum equivalent hydrogen consumption

This strategy converts the electric energy of the power battery into fuel consumption. Without considering the internal resistance of the power battery and the energy loss during the charging and discharging process, the equivalent fuel consumption C bat of the battery during charging and discharging is:

Where C st,svg is the average fuel consumption rate of the fuel cell; P st,avg is the average output power of the fuel cell; P bat is the instantaneous power of the battery. The equivalent fuel consumption function of the system is:

Csys = Cst + γ (SOC) Cbat

In the formula, γ is the SOC compensation coefficient, which is used to ensure that the SOC of the high-voltage power battery is within a reasonable range. Its value is:

SOC H is the given upper limit of SOC; SOC L is the given lower limit of SOC; β is the equivalent factor (0 < β < 1). The system equivalent fuel consumption Jmin is taken as the optimization target:

The constraints for setting the optimization problem are as follows:

• The sum of the output power of the fuel cell and the output power of the high-voltage power battery is equal to the power demand of the entire vehicle;

• The output power of the high-voltage power battery is within the output range determined by its current SOC;

• The output power of the fuel cell is within the maximum and minimum values ​​allowed;

• The output power change rate of the fuel cell system meets the requirements of the stack system supplier (such as 34Kw/S);

• The minimum operating power of the fuel cell system must meet the requirements of the fuel cell system supplier (such as 24Kw);

In the ECMS control strategy, the role of the equivalent factor is to adjust the speed of adjusting the SOC of the high-voltage battery to the target value. If the equivalent factor is too large, it will cause excessive power consumption when the power battery SOC is high. When the SOC drops quickly, the hydrogen consumption will increase greatly for charging, and finally increase the hydrogen fuel consumption. If the equivalent factor is too small, the charging will be too slow after the SOC is reduced, which may cause the SOC to be too low, affecting the output efficiency and service life of the battery. Therefore, the formulation of the equivalent factor is closely related to the SOC of the high-voltage power battery. Therefore, the high-voltage battery SOC should be considered when formulating the equivalent factor. When the SOC is very high, the equivalent factor should be increased to utilize the spare energy of the battery. When the SOC is very low, the equivalent factor should also be increased to quickly maintain the SOC of the high-voltage power battery. When the SOC is within a reasonable range, the equivalent factor should be reduced, which can better maintain the SOC and reduce the equivalent factor. Therefore, the equivalent factor β is set

When SOCH and SOCL are 0.8 and 0.2 respectively, the optimal equivalent factor changes with SOC as shown in the figure below. It can be seen that when the battery SOC is close to SOCH or SOCL, the optimal equivalent factor β is close to 1; when the battery SOC is close to (SOCH + SOCL)/2, the optimal equivalent factor β approaches 0, which has the same characteristics as the ideal optimal equivalent factor in the above analysis.

Optimal equivalent factor-power battery SOC curve

3.2 Braking Energy Recovery Optimization

The main steps of the braking energy recovery control strategy are: during braking, the brake controller (such as EBS) calculates the total braking force required by the driver according to the driver's brake pedal stroke, distributes the braking force to the front and rear axles on the premise of meeting the braking intensity requirements and braking stability, and then distributes the braking force allocated to the rear axle to electric braking and mechanical braking. The principle is to use electric braking as much as possible to reduce the use of mechanical braking. When the motor braking does not meet the braking force requirements, the mechanical braking will compensate to ensure the smooth deceleration of the vehicle. At the same time, the connection process between electric braking and mechanical braking cannot affect the driver's braking feeling. In addition, if the anti-lock braking ABS or the vehicle stability control ESC related functions are activated during the electric braking process, it is necessary to exit the electric braking and the mechanical braking control will be performed by EBS.

3.2.1 System structure of regenerative braking

As shown in the figure below, the regenerative braking system mainly includes the following components:

• Brake valve (foot valve): adopts electric and pneumatic dual circuit design, and monitors the driver's braking intention through dual pedal displacement sensors;

• Axle control module: divided into front axle control module, rear axle control module, middle axle control module (equipped for 3-axle vehicles), and the axle hole module can control the brake air pressure of the left and right wheels of the corresponding axle through the solenoid valve. It has two working modes: electric control mode and air control mode to achieve redundant backup. When the electric control mode fails, the system enters the air control state;

• Air reservoir: air is pumped into the air reservoir through an air pump to maintain the air pressure of the brake system;

• Brake (spring cylinder): a device that generates braking force at the wheel end;

• Wheel speed sensor: detects the wheel speed of each wheel and is connected to the EBS controller for road adhesion coefficient identification, etc.

• Electronic brake controller EBS: As the main control module of braking, it collects signals from various sensors, calculates and distributes braking force, and coordinates electric braking and mechanical braking, etc.

• Motor/electronic control: During the braking process, the motor is in power generation mode, providing reverse braking torque;

• High-voltage battery, etc.: Stores electrical energy recovered during braking.

Regenerative braking system structure

3.2.2 Regenerative braking control strategy

1. Front and rear braking force distribution

The front and rear braking forces are mainly distributed according to the ideal braking force distribution curve, GB12676 standard and the deceleration distribution of common cycle conditions. Under the premise of ensuring braking stability, the braking force is distributed to the rear wheels as much as possible. According to automobile theory, the ideal braking force distribution curve and the simultaneous locking of the front and rear wheels are more beneficial to the utilization of road adhesion conditions and directional stability during braking.

In order to achieve maximum braking energy recovery and take into account braking stability, the braking intensity under common working conditions is considered to be below a certain calibration value (such as 0.15). Therefore, when the braking intensity is below 0.15, the braking force is fully distributed to the rear wheels. When the braking intensity is greater than 0.15, the front wheel mechanical brake is gradually involved. Based on the above principles, the braking process is roughly divided into the following 3 sections:

• 0A stage: At this time, the braking intensity is reduced, and the braking force is fully distributed to the rear wheels, and the braking is performed through motor feedback;

• AB section: Its braking intensity is relatively high, and the front and rear wheels need to provide braking force together;

• After section B: the braking force of the front and rear axles is distributed according to the ideal braking distribution curve (I curve);

2. Calculation of motor feedback torque

The maximum regenerative torque of the motor is limited by the current motor speed, motor temperature, high-voltage system status, high-voltage power battery SOC and charging power. The following factors should be considered when calculating the motor's braking torque:

• The maximum voltage and feedback current of the motor feedback cannot exceed the current instantaneous charging current limit of the high-voltage power battery. Usually, the rechargeable current of the battery will decrease when the SOC is high. Therefore, in order to improve the efficiency of brake energy recovery, the SOC of the high-voltage power battery is usually avoided to be too high. However, as described in Section 3.2 above, the high-voltage power battery needs to play the role of "peak shaving and valley filling" in dynamic conditions in hydrogen fuel cell heavy truck models, and the battery SOC cannot be too low. How to dynamically adjust the target SOC of the high-voltage battery according to different driving conditions will be a complex and important task. For example, combining the high-precision map information of the high-level intelligent driving system to predict the topology of the road ahead, road traffic conditions, etc., appropriately increase the target SOC of the power battery before going uphill, and appropriately reduce the power battery SOC before a long downhill, so as to maximize the recovery of brake energy;

• The feedback torque limit of the current motor. EBS needs to receive the feedback torque limit of each motor of each middle axle and rear axle, so that the torque allocated to the electric brake cannot exceed the torque limit of the current motor;

• When ABS or ESC is activated, brake energy recovery must be exited. In emergency conditions such as ABS activation or ESC activation, electric braking cannot meet real-time requirements and needs to be controlled by mechanical braking;

• When a distributed drive is used, torque coordination and distribution between distributed motors need to be considered during braking.

3.3 Optimization of motor torque distribution based on efficiency

Under the same driving conditions, the drive system needs to output the same speed and torque, that is, the output power is consistent. The significance of torque optimization distribution is to improve the efficiency of the drive system, thereby reducing the input power of the battery. Then the optimal distribution of torque can be simplified to the optimal solution of the torque distribution coefficient λ between the front and rear axles:

λ = Tf /Td, Td = Tf + Tr

Where: λ is the distribution coefficient of the front axle torque. When the front motor is used as the main drive motor, 0.5 ≤λ≤1; Tf is the torque required by the front axle (N·m); Tr is the torque required by the rear axle (N·m); Td is the total torque requirement (N·m). The efficiency map of the selected motor is shown in the figure below.

Drive Motor Efficiency Map

At the same time, the motor output efficiency η is regarded as a function η(T, n) of the motor torque T and the speed n. The energy utilization rate of the drive system is

In summary, the mathematical model for the optimal efficiency of the drive system is:

The constraints are

Where T max is the maximum torque that the intermediate shaft motor can provide. The above formula represents the limitation of the maximum output characteristics that the motor can provide. Using Matlab to perform numerical solution, we can obtain the optimal efficiency of λ with respect to torque T and speed n.

The distribution coefficient is shown below

From the above figure, we can see that the distribution coefficient is most widely distributed in the area between 0.5 and 1, which means that under low torque demand, a single motor can provide the power required for the vehicle to travel. At this time, λ = 1. When the power demand is large, the torque demand of the whole vehicle is evenly distributed between the front and rear motors (the front and rear motors are the same). At this time, λ = 0.5, which can avoid a single motor working in an inefficient area at the edge.

For power systems with distributed drive, such as three motors (middle axle integrated bridge + rear axle distributed bridge) and four motors (middle axle and rear axle distributed bridges), in addition to considering the above-mentioned motor torque distribution based on optimal economy, the torque distribution based on stability is also considered. On high-adhesion system roads, the motor efficiency is maximized by optimizing the motor torque distribution. On low-adhesion coefficient roads, the motor torque distribution based on stability is adopted to ensure that the wheel slip rate is close to the optimal slip rate.

Schematic diagram of torque control based on stability

3.4 Thermal management optimization based on waste heat recovery

The thermal management system of fuel cell vehicles mainly integrates the thermal management of the cab, fuel cell, power battery, motor and high-voltage accessories, and manages them in a unified and coordinated manner, changing the original relatively independent management mode. Since heat pump air conditioners are highly efficient, energy-saving, and have integrated hot and cold regulation, a vehicle thermal management solution can be designed based on the heat pump air conditioning system, which can cool the cab and dissipate heat for fuel cells and power batteries in summer, heat the cab, preheat and keep warm the fuel cells and power batteries in winter, and reuse the waste heat of the battery pack and motor.

Schematic diagram of the thermal management system of a fuel cell vehicle

The principle of the thermal management system is shown in the figure above. The vehicle thermal management system consists of four circulation loops, namely, the heat pump air conditioning circulation loop, the fuel cell circulation loop, the power battery circulation loop, and the motor cooling circulation loop. There is a four-way reversing valve and five solenoid valves in the heat pump air conditioning circulation loop to adjust the circulation path and direction of the refrigerant in the loop, and the refrigerant flow rate in the loop is adjusted by the electric compressor. The fuel cell loop and the power battery circulation loop adjust the refrigerant flow rate in the loop through water pump 1 and water pump 2 respectively. There is a three-way solenoid valve in the motor cooling loop to adjust the path of the refrigerant in the loop, and the refrigerant flow rate in the loop is adjusted by water pump 3. The heat pump air conditioning circulation loop exchanges heat with the fuel cell circulation loop and the power battery circulation loop in the heat pump heat exchanger; the motor cooling circulation loop exchanges heat with the cab through the in-vehicle heat exchanger. Since the motor only needs cooling but not preheating, the motor cooling circulation loop is set separately and does not exchange heat with the heat pump air conditioning circulation loop. Each circulation loop achieves the purpose of temperature regulation by adjusting the coolant flow rate.

The working mode of the vehicle thermal management system is adjusted through the logic threshold control strategy, collecting real-time ambient temperature Tamb , cab temperature Tcab , fuel cell temperature Tful , power battery temperature Tpow , and motor temperature Tm . The logical relationship of the logic threshold control state is shown in the figure below.

Thermal management system logic threshold logic relationship block diagram

According to the logic threshold control, the vehicle thermal management system is divided into the following specific working modes:

• Mode 1 Cab cooling only: Ambient temperature Tamb > 20°C, cab temperature Tcab > 25°C, fuel cell temperature 80°C > Tful > 70°C, power battery temperature 45°C > Tpow > 18°C. The four-way reversing valve is turned to the cooling position, solenoid valve 5 is opened, and the remaining four battery valves are closed.

• Mode 2 Cab cooling, battery pack heat dissipation: Ambient temperature Tamb > 20°C, cab temperature Tcab > 25°C, fuel cell temperature Tful > 80°C, power battery temperature Tpow > 45°C. The four-way reversing valve is turned to the cooling position, solenoid valves 2, 4 and 5 are opened, and solenoid valves 1 and 3 are closed.

• Mode 3 Cab cooling, battery pack preheating: Ambient temperature Tamb > 20 °C, cab temperature Tcab > 25 °C, fuel cell temperature Tful < 70 °C, power battery temperature Tpow < 18 °C. The four-way reversing valve is turned to the cooling position, solenoid valves 1 and 3 are opened, and solenoid valves 2, 4, and 5 are closed.

• Mode 4 Cab heating only: ambient temperature Tamb < 10 ℃, cab temperature Tcab < 18 ℃, fuel cell temperature 80 ℃ > Tful > 70 ℃, power battery temperature 45 ℃ > Tpow > 18 ℃. The four-way reversing valve is turned to the heating position, solenoid valve 5 is opened, and the other four battery valves are closed.

• Mode 5 Cab heating, battery pack preheating: Ambient temperature Tamb ＜ 10 ℃, cab temperature Tcab ＜ 18 ℃, fuel cell temperature Tful ＜ 70 ℃, power battery temperature Tpow ＜ 18 ℃. The four-way reversing valve is turned to the heating position, solenoid valves 2, 4 and 5 are opened, and solenoid valves 1 and 3 are closed.

• Mode 6 Cab heating, battery cooling: Ambient temperature Tamb < 10°C, cab temperature Tcab < 18°C, fuel cell temperature Tful > 80°C, power battery temperature Tpow > 45°C. The four-way reversing valve is turned to the heating position, solenoid valves 1 and 3 are opened, and solenoid valves 2, 4, and 5 are closed.

• Mode 7 Battery pack heat source defrosting mode: Ambient temperature Tamb < 10°C, fuel cell temperature Tful > 80°C, power battery temperature Tpow > 45°C, and frosting on the external heat exchanger. The four-way reversing valve is turned to the cooling position, solenoid valves 2 and 4 are opened, and solenoid valves 1, 3, and 5 are closed.

• Mode 8 Cabin heat source defrosting mode: Ambient temperature Tamb < 10 ℃, fuel cell temperature Tful < 70 ℃, power battery temperature Tpow < 18 ℃, and frosting on the external heat exchanger. The four-way reversing valve is turned to the cooling position, only solenoid valve 5 is opened, and the other four solenoid valves are closed.

• Mode 9 Battery pack independent thermal management, battery pack independent heat dissipation: Open solenoid valves 1 and 4, close solenoid valves 2, 3 and 5. When the ambient temperature is 20 ℃ ＞ T amb ＞ 10 ℃, the cab does not need cooling or heating; the fuel cell temperature T ful ＞ 80 ℃, the power battery temperature T pow ＞ 45 ℃, enter the battery pack heat dissipation condition; the fuel cell temperature T ful ＜70 ℃, the power battery temperature T pow ＜ 18 ℃, enter the battery pack preheating condition.

• Mode 10 Motor cooling mode: When the motor temperature Tm > 100℃, the motor enters cooling mode: When the ambient temperature Tamb < 10℃, the motor waste heat is utilized, otherwise the motor cools down alone.

3.5 Low-voltage system energy management optimization

1. DC-DC working mode optimization:

Under different working conditions of heavy trucks, the conversion efficiency of 24VDCDC is improved to make it work in the high-efficiency area as much as possible. For example, when the intelligent charging function is turned on, the output voltage of DCDC should be increased, and the high-power charging mode should be adopted to reduce the energy consumption of various high-voltage components of the vehicle during the charging period and shorten the charging time as much as possible. At the same time, during charging, the DCDC interval working mode can be adopted. Because the power of the low-voltage electrical appliances of the whole vehicle during charging is relatively small compared with the driving conditions, the interval working mode can make the DCDC work in the high-efficiency area as much as possible. When the SOC of the low-voltage battery is high, the DCDC output is stopped, and the power consumption of the low-voltage electrical appliances of the whole vehicle is supplied by the low-voltage battery as much as possible. When the SOC of the low-voltage battery is lower than the set threshold, the DCDC is started to supply power to the vehicle load and charge the battery.

2. Power distribution optimization

By using intelligent MOS to replace traditional fuses and relays, the wiring loops of the entire vehicle can be reduced and the wire diameter can be reduced. According to statistics, the wiring harness cost of the entire vehicle can be reduced by about 28% after the use of intelligent power distribution. At the same time, when intelligent power distribution is used, loads with no working requirements can be turned off based on different working conditions and scenarios of the vehicle, thereby reducing the low-voltage power consumption of the entire vehicle.

/END/