Power distribution method and system based on health state difference of multiple stacks of fuel cells

Abstract

The invention discloses a power distribution method and system based on fuel cell multi-stack health state difference, and belongs to the technical field of fuel cell application. The specific method comprises the following steps: monitoring operation parameters of a plurality of fuel cell stacks through a sensor; obtaining the health state of each fuel cell stack and the health state of the lithium battery; establishing a hydrogen consumption model, a fuel cell SOH loss model, a multi-stack SOH difference equivalent loss model, a lithium battery SOH loss model and a lithium battery start and end SOC loss model; the calculation loss is converted into equivalent hydrogen loss, and a target function model is constructed; and optimizing the target function by using a reinforcement learning algorithm, distributing power output of the fuel cell and the lithium battery according to the current required power and the health state through an energy management system, and regularly updating parameters of a parameter identification module and a model to realize dynamic optimization. The method can identify the life difference between multiple piles, balance the life difference to realize multi-pile performance balance and improve the overall life of a multi-pile system.

Description

Technical Field

[0001] The present invention relates to a power distribution method and system, in particular to a power distribution method based on the health state difference of multiple fuel cell stacks and a multi-stack fuel cell system, belonging to the technical field of fuel cell applications. Background Art

[0002] Currently, the large-scale use of fossil fuels has brought problems of environmental pollution and energy shortage, which has promoted the large-scale development and use of renewable energy. However, the intermittency and volatility of renewable energy production have led to spatio-temporal differences between the power generation side and the load side. To solve this problem, hydrogen energy is a long-term energy storage method that can span days, months, and seasons. That is, green hydrogen is produced by electrolyzing water when renewable energy is abundant, and fuel cells are used for power generation when renewable energy power generation is insufficient, while electricity demand is high and difficult to meet.

[0003] Fuel cells have many limitations such as limited single-stack power, reduced mass and heat transfer uniformity in high-power stacks, and high costs of auxiliary components. Therefore, multiple fuel cell stacks (Multi-Stack Fuel Cell, MSFC) are often used in parallel to achieve higher power output. MSFC can operate relatively independently between stacks through its modular design. Through reasonable control methods, faulty stacks can be isolated in a timely manner without affecting the normal operation of the entire system. Even replacing a faulty stack will not affect the overall electrical connection, thereby enhancing the reliability and maintainability of the power generation system. In addition, when a faulty stack group recovers its health state from adverse conditions such as flooding and membrane drying, it can be made to rejoin the power generation system through control strategies, improving the fault tolerance of the system. However, the introduction of multiple sets of fuel cells easily brings slow dynamic response of fuel cells and differences in operating performance between single stacks, which poses new problems for the control and management of multi-stack fuel cell systems. Slow dynamic response is an inherent characteristic of fuel cells. At the same time, rapid power response will also cause rapid decay of the fuel cell life. To solve this problem, an energy storage system such as a lithium battery or a super capacitor needs to be added to the fuel cell power generation system to suppress power fluctuations. At the same time, it is necessary to solve the problems of how to distribute power between multiple fuel cell stacks and the energy storage system according to the power generation characteristics of different power sources to achieve stable fuel cell output power, maintain the state of charge of the energy storage system, and the economy of system power generation.

[0004] The current technical solutions focus on optimizing the hydrogen consumption, life loss of multi-stack systems, as well as the life loss of lithium batteries and the equivalent hydrogen consumption of the SOC difference from start to end, in order to improve the overall economy, but ignore the differences in the health status among fuel cells. In the MSFC power generation system, some fuel cells have excellent output performance. Under the same output voltage, they can achieve higher output power and have a longer remaining service life; while some fuel cells can only achieve lower output power and have a shorter remaining service life. According to the "barrel theory", the remaining service life of the entire MSFC depends on the fuel cell with the lowest remaining service life. Therefore, in order to improve the overall life of the multi-stack system, it is necessary to develop a power distribution algorithm that can identify the life differences among multiple stacks, balance the life differences to achieve the performance balance of multiple stacks, and improve the overall life of the multi-stack system. Summary of the Invention

[0005] In order to solve the problem of the short service life of the multi-stack system, the present invention further provides a power distribution method and system based on the health status differences of multiple fuel cell stacks.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A power distribution method based on the health status differences of multiple fuel cell stacks, the power distribution method based on the health status differences of multiple fuel cell stacks is realized through the following steps:

[0008] S1: Monitor the operating parameters of multiple fuel cell stacks through sensors, and the operating parameters include stack temperature, anode and cathode inlet pressures, anode and cathode flow channel humidities, and anode and cathode gas partial pressures;

[0009] S2: Based on the monitoring data obtained in S1, obtain the health status of each fuel cell stack through a parameter identification module;

[0010] S3: Obtain the health status of the lithium battery according to the state of charge of the lithium battery;

[0011] S4: Based on the health status of the fuel cell stack and the health status of the lithium battery obtained in S2 and S3, establish a hydrogen consumption model, a fuel cell SOH loss model, a multi-stack SOH difference equivalent loss model, a lithium battery SOH loss model, and a lithium battery start-end SOC loss model;

[0012] S5: Convert the losses calculated by the models established in S4 into equivalent hydrogen losses, and construct an objective function model;

[0013] S6: Use the reinforcement learning algorithm Proximal Policy Optimization to optimize the objective function, and train the model based on historical operation data to obtain an optimized power distribution strategy;

[0014] S7: Optimize the power distribution strategy obtained in S6, and through the energy management system, allocate the power outputs of the fuel cell and the lithium battery according to the current required power and the health state, and output the reference current of the DC / DC to achieve power distribution;

[0015] S8: Regularly update the parameter identification module and the model parameters to achieve dynamic optimization.

[0016] Furthermore, the health state of the fuel cell stack is calculated by the following formula:

[0017]

[0018] where, v 0 is the voltage of the fuel cell stack when not in use, v fc,run is the current operating voltage, v fc,rated is the rated voltage of the current operation, v rated is the rated voltage obtained through polarization curve identification; v fc = E - (v ohm + v act + v conc ) is the output voltage of the PEMFC, E is the open-circuit voltage of the PEMFC, v ohm is the ohmic loss voltage, v act is the activation loss voltage, v conc is the concentration difference loss voltage.

[0019] Furthermore, the hydrogen consumption model is calculated by the following formula:

[0020]

[0021] where, N cell is the number of single cells in the fuel cell stack, is the molar mass of hydrogen, F is the Faraday constant, λ is the excess ratio of hydrogen, C fc is the hydrogen consumption rate; P fc,run is the power during the operation of the fuel cell.

[0022] Furthermore, the fuel cell SOH loss model is divided into start-stop, idle, low-power constant load, medium-power constant load, high-power constant load, rapid load change, and slow load change according to seven working conditions, and is calculated by the following formula:

[0023] The voltage loss at each moment is:

[0024] ΔU(t) = F(t) * (A * Δt 1 + B * Δt 2 + C * Δt 3 + D * Δt 4+E*ΔP 1 +F*ΔP 2 +G*α)

[0025] Among them, F(t) is a coefficient that varies with the operating time of the fuel cell, A, B, C, D, E, F, and G are coefficients corresponding to seven working conditions respectively, Δt is the operating time, ΔP is the power change, and α is the number of times;

[0026] The SOH loss of the stack is:

[0027]

[0028] Among them, p Stack is the price coefficient of the fuel cell stack, P rated is the rated voltage of the stack operation,

[0029] The SOH change of the stack is:

[0030]

[0031] Furthermore, the multi-stack SOH difference equivalent loss model is calculated by the following formula:

[0032]

[0033] Among them, SOH i,min is the stack with the shortest life in the multi-stack electric cone, β is the price coefficient related to the life loss of the fuel cell, and n is the number of stacks in the multi-stack electric cone.

[0034] Furthermore, the lithium battery SOH loss model is calculated by the following formula:

[0035] The lithium battery adopts a common second-order RC equivalent circuit model, and its life loss is:

[0036]

[0037] Among them, |I(t) is the current of the lithium battery operation, Δt is the time constant, N(c,T a ) is the number of cycles related to the battery temperature, C n is the battery capacity.

[0038] The lithium battery SOH equivalent loss is:

[0039]

[0040] Furthermore, the lithium battery initial and final SOC loss model is calculated by the following formula:

[0041] C BAT,ele =γ×(SOC beg -SOCend )

[0042] Among them, γ is a coefficient related to the electricity price and the hydrogen price, SOC beg is the state of charge at the start, SOC end is the state of charge at the end.

[0043] Furthermore, the objective function is calculated by the following formula:

[0044]

[0045] Among them, w FC,i is the weight coefficient related to the fuel cell price, w diff is the weight coefficient related to the difference in the health state of the fuel cell, w BAT is the weight coefficient related to the life loss of the lithium battery, w soc is the weight coefficient related to the SOC loss of the lithium battery.

[0046] Furthermore, the state of the Proximal Policy Optimization of the reinforcement learning algorithm is the current demand power and the health state, the action is the power distribution strategy of the fuel cell and the lithium battery, and the reward is the output of the objective function.

[0047] Furthermore, the energy management system outputs the reference current of the DC / DC according to the optimized power distribution strategy, and realizes dynamic optimization and system life maximization by regularly updating the parameter identification module and the model parameters.

[0048] A fuel cell multi-stack system for implementing the power distribution method based on the difference in the health state of multiple fuel cells, the fuel cell multi-stack system includes:

[0049] Multiple parallel fuel cell stacks, and the fuel cells are connected to the DC bus through a boost DC / DC controller;

[0050] A lithium battery, and the lithium battery is connected to the DC bus through a bidirectional DC / DC;

[0051] Multiple fuel cell system sensors for monitoring the operating parameters of the stacks;

[0052] Multiple fuel cell parameter identification modules for obtaining the health state of the stacks;

[0053] A lithium battery parameter identification module for obtaining the health state of the lithium battery;

[0054] An energy management system for executing the power distribution strategy.

[0055] Furthermore, the system delays the overall lifespan of the multi-stack system and reduces hydrogen consumption, lithium battery loss, and power loss by dynamically optimizing the power distribution strategy.

[0056] The beneficial effects of the present invention are as follows:

[0057] 1. The present invention can achieve relatively small hydrogen consumption, single-stack lifespan loss, and lithium battery lifespan loss in the fuel cell multi-stack system, optimize the difference in the health status between the stacks, and thus extend the remaining service life of the entire multi-stack system.

[0058] 2. The present invention can achieve the power distribution effect with the maximum overall lifespan of the fuel cell multi-stack system under the condition that the sum of hydrogen consumption, fuel cell stack lifespan loss, lithium battery lifespan loss, and power loss of the fuel cell is relatively small. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a schematic structural diagram of an embodiment of the fuel cell multi-stack system implementing the power distribution method based on the health status difference of the fuel cell multi-stack of the present invention;

[0060] Figure 2 is a schematic flowchart of the power distribution method based on the health status difference of the fuel cell multi-stack of the present invention;

[0061] Figure 3 is a schematic diagram comparing the SOH changes of the fuel cells in the multi-stack system. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] DETAILED DESCRIPTION OF THE EMBODIMENT 1: In combination with Figures 1 - 3 to illustrate this embodiment. As Figure 1 shown, the fuel cell multi-stack system implementing the power distribution method based on the health status difference of the fuel cell multi-stack in this embodiment includes: a plurality of fuel cell stacks connected in parallel, the fuel cells are connected to the DC bus through a boost DC / DC controller; a lithium battery, the lithium battery is connected to the DC bus through a bidirectional DC / DC; a plurality of fuel cell system sensors for monitoring the operating parameters of the stacks; a plurality of fuel cell parameter identification modules for obtaining the health status of the stacks; a lithium battery parameter identification module for obtaining the health status of the lithium battery; and an energy management system for implementing the power distribution strategy. The system delays the overall lifespan of the multi-stack system and reduces hydrogen consumption, lithium battery loss, and power loss by dynamically optimizing the power distribution strategy.

[0063] Specifically, three fuel cells are connected in parallel to form a multi-stack system. The fuel cells are connected to the DC bus through a boost DC / DC controller, and the lithium battery is connected to the DC bus through a bidirectional DC / DC. The fuel cell system sensors monitor the stack temperature T, the anode and cathode inlet pressures P c and Pa and the cathode and anode flow channel humidity RH c and RH a and the cathode and anode gas partial pressures and The sensor information is transmitted to the parameter identification module to obtain the corresponding health state SOH of each fuel cell stack. The lithium battery parameter identification module obtains the health state SOH of the lithium battery according to the state of charge SOC BAT . Finally, the energy management system, based on the SOH of the fuel cell and the SOH of the lithium battery BAT and the current required power P need , allocates the reference current i fc and i BAT .

[0064] As Figure 2 shown, for the power allocation method based on the health state differences of multiple fuel cell stacks described in this embodiment, the power allocation method based on the health state differences of multiple fuel cell stacks is implemented through the following steps:

[0065] S1: Monitor the operating parameters of multiple fuel cell stacks through sensors, where the operating parameters include stack temperature, cathode and anode inlet pressures, cathode and anode flow channel humidity, and cathode and anode gas partial pressures;

[0066] S2: Based on the monitoring data obtained in S1, obtain the health state of each fuel cell stack through the parameter identification module;

[0067] The fuel cell polarization curve model can reflect the evolution law of the fuel cell output performance with the change of the health state. However, due to the numerous non-linear parameters in the polarization curve model, identification is required:

[0068] v fc = E - (v ohm + v act + v conc ) (1)

[0069] In formula (1), v fc is the output voltage of the PEMFC, E is the open circuit voltage of the PEMFC, v ohm is the ohmic loss voltage, v act is the activation loss voltage, v conc is the concentration difference loss voltage. These variables are related to the stack temperature T, the cathode and anode inlet pressures P c and P a , the cathode and anode flow channel humidity RH c and RH a , the cathode and anode gas partial pressures and .

[0070] The state of health (SOH) of a fuel cell stack is calculated by the following formula:

[0071]

[0072] In Equation (2), v 0 is the voltage of the fuel cell stack when not in use, v fc,run is the current operating voltage, v fc,rated is the rated voltage during the current operation, v rated is the rated voltage obtained by identifying the polarization curve.

[0073] Specifically, based on the stack temperature T measured by the stack sensor, the anode and cathode inlet pressures P c and P a the anode and cathode flow channel humidities RH c and RH a the anode and cathode gas partial pressures and the SOH of the stack is obtained according to Equations (1) and (2).

[0074] S3: Obtain the state of health of the lithium battery according to the state of charge of the lithium battery;

[0075] S4: Based on the state of health of the fuel cell stack and the state of health of the lithium battery obtained in S2 and S3, establish a hydrogen consumption model, a fuel cell SOH loss model, a multi-stack SOH difference equivalent loss model, a lithium battery SOH loss model, and a lithium battery initial and final SOC loss model;

[0076] Hydrogen consumption model considering the difference in the state of health of multiple stacks:

[0077] The hydrogen consumption of the fuel cell stack is:

[0078]

[0079] In Equation (3), N cell is the number of single cells in the fuel cell stack, is the molar mass of hydrogen, F is the Faraday constant, λ is the excess ratio of hydrogen, C fc is the hydrogen consumption rate; P fc,run is the power during the operation of the fuel cell.

[0080] From Equation (2), it can be obtained that:

[0081]

[0082] Combining Equation (3), the calculation formula of the hydrogen consumption model is as follows:

[0083]

[0084] As can be seen from Equation (5), in a multi-stack system, the health differences between multiple stacks will result in different hydrogen consumption curves for multiple stacks. Therefore, the energy management system needs to fully consider the system efficiency differences brought about by different hydrogen consumption curves.

[0085] Fuel cell SOH loss model:

[0086] According to the load demand, seven operating conditions are divided: start-stop, idle, low-power constant load, medium-power constant load, high-power constant load, rapid load change, and slow load change. The voltage loss at each moment is:

[0087] ΔU(t) = F(t) * (A * Δt 1 + B * Δt 2 + C * Δt 3 + D * Δt 4 + E * ΔP 1 + F * ΔP 2 + G * α) (6)

[0088] In Equation (6), F(t) is a coefficient that changes with the operating time of the fuel cell, A, B, C, D, E, F, and G are coefficients corresponding to the seven operating conditions respectively, Δt is the operating time, ΔP is the power change, and α is the number of times.

[0089] The SOH loss of the stack is:

[0090]

[0091] Among them, p Stack is the price coefficient of the fuel cell stack, P rated is the rated voltage of the stack operation,

[0092] The SOH change of the stack is:

[0093]

[0094] Multi-stack SOH difference equivalent loss model:

[0095] The multi-stack SOH difference equivalent loss model is calculated by the following formula:

[0096]

[0097] Among them, SOH i,min is the stack with the shortest life in the multi-stack stack, β is the price coefficient related to the life loss of the fuel cell, and n is the number of stacks in the multi-stack stack.

[0098] Since the life of the multi-stack depends on the SOH of the stack with the shortest life i,minThe lifespan, so the sum of the differences in SOH of the three stacks compared to SOH can be used as the loss. i,min is used as the loss.

[0099]

[0100] Lithium battery SOH loss model:

[0101] The lithium battery adopts a common second-order RC equivalent circuit model, and its life loss is:

[0102]

[0103] Among them, |I(t) is the current of the lithium battery during operation, Δt is the time constant, N(c,T a ) is the number of cycles related to the battery temperature, C n is the battery capacity.

[0104] The equivalent loss of the lithium battery SOH is:

[0105]

[0106] Lithium battery initial and final SOC loss model:

[0107] During the operation of the lithium battery, the SOC at the start and end is different in most states, which means that there is power loss or replenishment in the lithium battery. Therefore, this part also needs to be considered:

[0108] C BAT,ele =γ×(SOC beg -SOC end ) (12)

[0109] Among them, γ is the coefficient related to the electricity price and hydrogen price, SOC beg is the SOC value at the start, and SOC end is the SOC value at the end.

[0110] S5: Convert the loss calculated by the model established in S4 into equivalent hydrogen loss to construct the objective function model;

[0111] Objective function model:

[0112] The objective function of the power distribution method considering the health differences of multiple stacks is:

[0113]

[0114] Among them, w FC,i is the weight coefficient related to the fuel cell price, w diff is the weight coefficient related to the health state difference of the fuel cell, w BATis the weight coefficient related to the life loss of the lithium battery, w soc is the weight coefficient related to the SOC loss of the lithium battery.

[0115] That is, the SOH identified according to different stack parameters i Establish a hydrogen loss model, a life loss model, and a multi-stack SOH difference equivalent loss model, as well as a lithium battery SOH loss model and a lithium battery initial and final SOC loss model. Multiply the losses calculated by the five models by the corresponding coefficients and add them together to obtain the objective function model.

[0116] S6: Use the reinforcement learning algorithm Proximal Policy Optimization to optimize the objective function, train the model based on historical operation data, and obtain the optimized power distribution strategy;

[0117] The state of the reinforcement learning algorithm Proximal Policy Optimization is the current required power and the health state, the action is the power distribution strategy of the fuel cell and the lithium battery, and the reward is the output of the objective function.

[0118] This embodiment uses the reinforcement learning algorithm Proximal Policy Optimization (PPO), trains based on the historical operation data of multiple stacks, then retains the optimized algorithm parameters, and actually runs on the multi-stack energy management platform. According to the parameter identification results, the objective function model is updated regularly. Among them, the state and action of the algorithm are:

[0119] state = {SOC, SOH BAT , SOH 1 , P 1 , SOH 2 , P 2 , SOH 3 , P 3 , P need} (14)

[0120] action = {P 1 , P 2 , P 3} (15)

[0121] S7: Distribute the optimized power distribution strategy obtained in S6 through the energy management system according to the current required power and the health state to allocate the power output of the fuel cell and the lithium battery, and output the reference current of the DC / DC to achieve power distribution;

[0122] That is, the output of the objective function is used as the reward of the Proximal Policy Optimization (PPO) algorithm, and the state is {SOC, SOH BAT , SOH 1 , P 1 , SOH 2 , P 2 , SOH 3 , P 3 , P need}, and the action is {P 1 , P 2 , P 3}. Using the multi-stack historical demand power data as the training set, the converged parameters are obtained, and the optimized algorithm is run on the energy management platform.

[0123] S8: Regularly update the parameter identification module and model parameters to achieve dynamic optimization; regularly update the SOH of each stack according to the sensor data parameter identification.

[0124] As Figure 3 shown in the left (a), a multi-stack system is composed of three 60kW fuel cells in parallel. When the multi-stack power distribution method considering the difference in health status of the present invention is adopted, the SOH changes of the three fuel cell stacks are close, which helps to delay the life of this multi-stack system. As Figure 3 shown in the right (b), when the difference in health status between multiple stacks is not considered, the SOH of a certain stack in the entire multi-stack system decays significantly faster, which reduces the life of the entire multi-stack system.

[0125] Working principle:

[0126] First, a hydrogen consumption model considering the differences in the health states of multiple stacks needs to be established based on the information from the sensors and the identified health state of the fuel cell. Then, the required power is divided into seven types: start / stop, idle, low-power constant load, medium-power constant load, high-power constant load, rapid load change, and slow load change, and a single-stack SOH loss model of the fuel cell is established accordingly. To balance the SOH differences between multiple stacks, an equivalent loss model for the SOH differences of multiple stacks is established. At the same time, a lithium battery life loss model and an electricity consumption model from start to end are established, and based on the prices of lithium batteries, hydrogen, and fuel cells, the losses calculated by the above models are converted into equivalent hydrogen losses, thereby establishing the objective function and constraints. Then, the Proximal Policy Optimization (PPO) algorithm is used to train with the historical operation data of the multi-stack system as the training set, and the trained model is applied to the energy management system of the multi-stack system. The energy management system outputs the reference current of the DC / DC according to the current health states of the fuel cell and the lithium battery and the power demand. Finally, after a period of operation, the parameter identification module updates the corresponding parameters, and the hydrogen consumption model, the fuel cell SOH loss model, the lithium battery life loss model, and the electricity consumption model from start to end are updated accordingly.

[0127] As described above, it is only a preferred embodiment of the present invention and does not impose any formal limitations on the present invention. Although the present invention has been disclosed above with the preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or equivalents by using the disclosed technical content within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention and is based on the technical essence of the present invention, any simple modification, equivalent replacement, and improvement made to the above embodiments still fall within the protection scope of the technical solution of the present invention.

Claims

1. A power allocation method based on the difference in health status of multiple fuel cell stacks, characterized in that: The power allocation method based on the difference in health status of multiple fuel cell stacks is implemented by the following steps: S1: monitoring operating parameters of multiple fuel cell stacks through sensors, wherein the operating parameters include stack temperature, cathode and anode air intake pressure, cathode and anode flow channel humidity, and cathode and anode gas partial pressure; S2: Based on the monitoring data obtained in S1, the health status of each fuel cell stack is obtained through the parameter identification module; S3: Obtaining a health status of the lithium battery according to the state of charge of the lithium battery; S4: Based on the health status of the fuel cell stack and the health status of the lithium battery obtained in S2 and S3, a hydrogen consumption model, a fuel cell SOH loss model, a multi-stack SOH difference equivalent loss model, a lithium battery SOH loss model, and a lithium battery initial and final SOC loss model are established; S5: convert the loss calculated by the model established in S4 into equivalent hydrogen loss and construct an objective function model; S6: Use the reinforcement learning algorithm Proximal Policy Optimization to optimize the objective function, train the model based on historical operation data, and obtain the optimized power allocation strategy; S7: The power distribution strategy optimized in S6 is used to distribute the power output of the fuel cell and the lithium battery according to the current power demand and health status through the energy management system, and the DC / DC reference current is output to achieve power distribution; S8: Regularly update the parameter identification module and model parameters to achieve dynamic optimization.

2. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 1 is characterized in that: The health status of the fuel cell stack is calculated by the following formula: Where v0 is the voltage of the fuel cell stack when it is not in use, v fc,run is the current operating voltage, v fc,rated is the rated voltage of the current operation, v rated is the rated voltage obtained through polarization curve identification; v fc =E-(v ohm +v act +v conc ) is the output voltage of PEMFC, E is the open circuit voltage of PEMFC, v ohm is the ohmic loss voltage, v act is the activation loss voltage, v conc is the concentration loss voltage.

3. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 2 is characterized in that: The hydrogen consumption model is calculated by the following formula: Among them, N cell is the number of cells in the fuel cell stack, is the molar mass of hydrogen, F is the Faraday constant, λ is the excess ratio of hydrogen, C fc is the consumption rate of hydrogen; P fc,run is the power during the operation of the fuel cell.

4. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 3 is characterized in that: The fuel cell SOH loss model is divided into seven operating conditions: start-stop, idling, low power constant load, medium power constant load, high power constant load, fast load change and slow load change, and is calculated by the following formula: The voltage loss at each moment is: ΔU(t)=F(t)*(A*Δt1+B*Δt2+C*Δt3+D*Δt4+E*ΔP1+F*ΔP2+G*α) Wherein, F(t) is the coefficient that changes with the operation time of the fuel cell, A, B, C, D, E, F, and G are the coefficients corresponding to the seven operating conditions, Δt is the operation time, ΔP is the power change, and α is the number of times; The SOH loss of the battery stack is: Among them, p Stack is the price coefficient of the fuel cell stack, P rated is the rated voltage of the stack, The SOH change of the battery stack is:

5. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 4 is characterized in that: The multi-stack SOH difference equivalent loss model is calculated by the following formula: Among them, SOH i,min is the shortest lifespan stack among the multiple stack cones, β is the price coefficient related to the life loss of the fuel cell, and n is the number of stacks in the multiple stack cones.

6. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 5 is characterized in that: The lithium battery SOH loss model is calculated by the following formula: Lithium batteries use the common second-order RC equivalent circuit model, and their life loss is: Among them, |I(t) is the current of the lithium battery, Δt is the time constant, N(c,T a ) is the number of cycles related to the battery temperature, C n is the battery capacity. The equivalent loss of lithium battery SOH is:

7. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 6 is characterized in that: The lithium battery SOC loss model is calculated by the following formula: C BAT,ele =γ×(SOC beg -SOC end ) Among them, γ is the coefficient related to electricity price and hydrogen price, SOC beg is the state of charge at the beginning, SOC end is the charge state at the end.

8. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 7 is characterized in that: The objective function is calculated by the following formula: Among them, w FC,i is the weight coefficient related to the fuel cell price, w diff is the weight coefficient related to the difference in fuel cell health status, w BAT is the weight coefficient related to the life loss of lithium battery, w soc is the weight coefficient related to the SOC loss of the lithium battery.

9. The power allocation method based on the difference in health status of multiple fuel cell stacks according to claim 8, characterized in that: The state of the reinforcement learning algorithm Proximal Policy Optimization is the current required power and health status, the action is the power allocation strategy of the fuel cell and the lithium battery, and the reward is the output of the objective function.

10. A fuel cell multi-stack system implementing the power allocation method based on the health status difference of fuel cell multi-stacks as described in any one of claims 1 to 9, characterized in that: The fuel cell multi-stack system comprises: Multiple fuel cell stacks connected in parallel, with the fuel cells connected to the DC bus via a boost DC / DC controller; Lithium battery, the lithium battery is connected to the DC bus via a bidirectional DC / DC; Multiple fuel cell system sensors to monitor stack operating parameters; Multiple fuel cell parameter identification modules to obtain the health status of the fuel cell stack; Lithium battery parameter identification module, used to obtain the health status of lithium batteries; Energy management system to implement power allocation strategy.

Images