Equalization charging method and system for multi-stack series-parallel system

Abstract

The invention relates to the field of energy storage, in particular to an equalizing charging method and system of a multi-stack series-parallel system. According to the method, the charging current distribution scheme is dynamically optimized by simulating the movement of particles in the search space, algorithm parameters can be automatically adjusted according to the real-time state of the electric pile, and the convergence speed and the global optimization capacity of the algorithm are improved. A health state evaluation index of the electric pile is introduced as a constraint condition of an APSO algorithm, parameters such as voltage, current, cycle index, capacity attenuation rate and internal resistance change of the electric pile are comprehensively considered, and excessive stress on the electric pile is avoided while equalizing charging is ensured. Therefore, the service life of the electric pile is prolonged, the maintenance cost of the system is reduced, and the charging efficiency and the safety of the whole system are improved. And a dynamic charging current adjusting method based on concentration distribution of electrolyte in the electric pile is introduced to further optimize the charging process, so that the safety and balance of charging are ensured.

Description

Technical Field

[0001] This invention relates to the field of energy storage, specifically to the field of charging control for multi-stack series-parallel systems, and to a method and system for equalizing charging of multi-stack series-parallel systems. Background Technology

[0002] Flow batteries, as a large-scale energy storage technology, are widely used in renewable energy storage, grid peak shaving and frequency regulation, and distributed energy systems. In these applications, battery stacks typically operate in a series-parallel configuration, making balanced charging control of the stacks crucial. Existing charging control methods for series-parallel battery stack systems are mainly based on fixed-ratio allocation or simple voltage and current feedback control. These methods typically employ preset allocation ratios or simple feedback adjustments based on the current voltage and current of the stacks when achieving current distribution among multiple stacks. For example, a common method is to allocate charging current according to a certain ratio based on the initial voltage level of the stacks, gradually bringing the voltages of each stack closer to uniformity. However, this method does not consider the dynamic changes of the stacks under different operating conditions, such as the stack's health status and capacity decay. Another method is to monitor the voltage and current of the stacks and adjust the charging current allocation ratio in real time to maintain the voltage and current of each stack within a certain range. While this method can adapt to dynamic changes in the stacks to some extent, it lacks a comprehensive assessment of the stack's health status, making precise current distribution difficult, and its control effect is not ideal when facing complex operating conditions.

[0003] Therefore, traditional methods struggle to achieve precise current distribution among multiple fuel cell stacks. Flow hydride fuel cell stacks are affected by various factors during operation, such as pipeline flow paths, cycle count, capacity decay, and electrolyte concentration variations. These factors lead to increasing performance differences among the stacks. Traditional methods cannot adjust the charging current distribution in real time based on these dynamic changes, easily resulting in some stacks being overcharged or undercharged, affecting stack lifespan and overall system performance. Furthermore, traditional methods cannot meet the charging requirements of different stacks under varying operating conditions. In practical applications, flow hydride fuel cell stacks may face different operating environments and load conditions, such as different depths of charge / discharge and temperature ranges. Traditional methods, lacking a comprehensive assessment of stack health status, cannot flexibly control charging according to the actual needs of the stacks, leading to low charging efficiency and potentially even safety hazards. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for equalizing charging a multi-fuel stack series-parallel system, which realizes equalizing charging control of the multi-fuel stack series-parallel system, improves the performance and reliability of the entire system, and extends the service life of the fuel stack.

[0005] To solve the aforementioned technical problem, the present invention adopts the following technical solution: a method for equalizing charging a multi-stack series-parallel system, comprising the following steps: S01. Data acquisition and preprocessing: Real-time acquisition of voltage, current, temperature, cycle count, capacity decay rate, and internal resistance change rate of each fuel cell stack, and preprocessing of the acquired data; S02. Stack health status assessment: The pre-processed capacity decay rate, internal resistance change rate and cycle number are weighted and summed to calculate the health status of each stack. S03. Optimize charging current distribution based on APSO algorithm, including: S31. Initialize the particle swarm size, maximum number of iterations, initial inertia weight and learning factor of the APSO algorithm, randomly generate the initial position of the particle swarm, and set the initial velocity of the particle swarm to 0. S32. Calculate the fitness value of each particle using the following formula: , in Let represent the fitness value of the i-th particle, and n represent the number of fuel cells. , Let represent the voltage and current of fuel cell j, respectively. , The average voltage and average current of all fuel cells are given. For the health state of the j-th fuel cell stack, Weighting coefficients for the health status of the fuel cell stack; S32. Update the optimal position and adjust the particle velocity and position; S33. Determine if the termination condition is met. If not, repeat steps S32 and S33. If it is met, output the optimal solution for charging current allocation. The optimal solution for charging current allocation is the globally optimal position of the particle swarm generated iteratively. S04. Charging control execution: Adjust the charging current of each battery stack according to the optimal scheme of charging current distribution to realize charging control of each battery stack.

[0006] Furthermore, in step S04, the charging current of the fuel cell stack is adjusted based on the electrolyte concentration distribution inside the stack. The adjustment method is as follows: The electrolyte concentration at different locations inside the fuel cell stack is monitored to obtain the electrolyte concentration difference within the stack. The charging current is then adjusted based on this concentration difference, using the following formula: , in The adjusted charging current, The charging current before adjustment. To adjust the coefficient, This represents the electrolyte concentration difference in the fuel cell stack.

[0007] Furthermore, in order to keep the total charging current constant, the charging current of each stack is redistributed based on the adjusted charging current. The distribution method is as follows: first, the difference between the adjusted total charging current and the original charging current is calculated, and then the difference is compensated to each charging stack according to the optimal scheme of charging current distribution calculated in step S03.

[0008] Furthermore, the electrolyte concentration at different locations inside the fuel cell stack is monitored in real time using an electrolyte concentration sensor.

[0009] Furthermore, electrolyte concentration sensors are distributed at the top, middle, and bottom of the fuel cell stack.

[0010] Furthermore, the optimal position is updated, and the particle velocity and position are adjusted as follows: if the fitness value of the current particle position is better than the fitness value of its individual optimal position, then the individual optimal position is updated to the current particle position; if the fitness value of a certain particle is better than the fitness value of the global optimal position, then the global optimal position is updated to the position of this particle. The formula for velocity update during iteration is: , The formula for velocity update during iteration is: , in Let i be the updated velocity of particle i. Let i be the velocity of particle i before the update. For inertial weights, , As a learning factor, , A random number within the interval [0,1]. For the optimal position of an individual, To be the globally optimal position This represents the position of particle i before the update. Let i be the updated position of particle i. Let be the velocity of particle i before the update.

[0011] Furthermore, step S05 involves real-time monitoring and adjustment, which involves real-time monitoring of the voltage, current, temperature, number of cycles, capacity decay rate, and internal resistance change rate of each fuel cell stack, and re-executing steps S02, S03, and S04 based on the real-time monitoring data.

[0012] This invention also discloses an equalization charging system for a multi-pile series-parallel system, comprising: A battery stack module includes multiple battery stacks, each consisting of multiple battery cells, for storing and releasing electrical energy; The data acquisition and preprocessing module is used to collect the voltage, current, temperature, number of cycles, capacity decay rate, and internal resistance change rate of each fuel cell stack in real time, and to preprocess the collected data. The health status assessment module evaluates the health status of the fuel cell stack based on the number of cycles, capacity decay rate, and internal resistance change rate after preprocessing. The APSO module employs an improved APSO algorithm to optimize the charging current allocation scheme based on the voltage, current, temperature, and health status of the fuel cell stack. This module finds the optimal charging current allocation scheme by simulating the movement of particles in the search space and automatically adjusts the APSO algorithm parameters according to the dynamic changes of the fuel cell stack, thereby improving the convergence speed and global optimization capability of the APSO algorithm. The charging control module controls the charging current of each battery stack according to the charging current distribution scheme.

[0013] Furthermore, it also includes a charging current adjustment module, which is used to adjust the charging current distribution scheme according to the electrolyte concentration difference inside the stack.

[0014] Furthermore, it also includes a reset module for performing a reset operation after charging is completed. The reset operation includes resetting the APSO parameters to restore the position and velocity of the particle swarm to the initial state; resetting the health status indicators of the health status assessment module to restore the health status indicators of each stack to the initial value; and resetting the charging current adjustment parameters of the charging control module to adjust the output current of the charging power supply to the initial value.

[0015] The beneficial effects of this invention are as follows: This invention achieves precise equal charging control for a multi-stack series-parallel battery system through an improved Adaptive Particle Swarm Optimization (APSO) algorithm, significantly improving system performance and reliability. Traditional methods have shortcomings in current distribution among multiple battery stacks, making it difficult to adapt to the dynamic changes of the stacks under different operating conditions, easily leading to overcharging or undercharging of some stacks, affecting service life and system performance. This invention, by simulating the movement of particles in the search space, dynamically optimizes the charging current distribution scheme, automatically adjusting algorithm parameters according to the real-time state of the stacks, improving the algorithm's convergence speed and global optimization capability.

[0016] This invention also introduces a health status assessment index for the fuel cell stack as a constraint condition for the APSO algorithm. It comprehensively considers parameters such as stack voltage, current, cycle count, capacity decay rate, and internal resistance changes to ensure that excessive stress on the stack is avoided while equalizing charging. This not only extends the lifespan of the fuel cell stack and reduces system maintenance costs, but also improves the charging efficiency and safety of the entire system.

[0017] The present invention also introduces a dynamic charging current adjustment method based on the electrolyte concentration distribution inside the fuel cell stack to further optimize the charging process and ensure the safety and balance of charging.

[0018] This invention can be applied to series-parallel systems of flow battery stacks of different specifications and quantities, as well as various application scenarios, such as renewable energy storage, grid peak shaving and frequency regulation, electric vehicle charging stations, and industrial energy storage systems. By monitoring the operating status of the stacks in real time and dynamically adjusting the charging control strategy based on real-time data, this invention enhances the system's response speed and adaptability, ensuring the efficient and safe operation of multi-stack series-parallel flow battery systems under complex operating conditions. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system described in Example 1; Figure 2 The flowchart is for the method described in Example 2. Figure 3 A flowchart for health status assessment; Figure 4 Flowchart for optimizing the sufficient current allocation scheme for the APSO algorithm; Figure 5 This is a flowchart for real-time monitoring and adjustment. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0021] Example 1 This embodiment discloses a leveling charging system for a multi-pile series-parallel system, such as... Figure 1 As shown, this system includes: The fuel cell stack module includes multiple fuel cell stacks, each of which includes multiple battery cells for storing and releasing electrical energy. Each fuel cell stack is equipped with a voltage acquisition module, a current sensor, and a temperature sensor for real-time monitoring of the fuel cell stack's electrical parameters and operating status.

[0022] The data acquisition and preprocessing module is used to collect the voltage, current, temperature, cycle count, capacity decay rate, and internal resistance change rate of each fuel cell stack in real time, and to preprocess and transmit the collected data to provide real-time data support for the APSO module.

[0023] The health status assessment module evaluates the health status of the fuel cell stack based on the number of cycles, capacity decay rate, and internal resistance change rate after preprocessing. This module calculates the health status indicators of each fuel cell stack by establishing a health status assessment model, providing a reference for charging control.

[0024] The APSO module, the core control module, employs an improved APSO algorithm to optimize the charging current allocation scheme based on the voltage, current, temperature, and health status of the fuel cell stack. This module finds the optimal charging current allocation scheme by simulating the movement of particles in the search space and automatically adjusts the APSO algorithm parameters according to the dynamic changes of the fuel cell stack, thereby improving the convergence speed and global optimization capability of the APSO algorithm.

[0025] The charging control module controls the charging current of each battery stack according to the optimized charging current distribution scheme of the APSO module. This module achieves precise charging control of each battery stack by adjusting the output current of the charging power supply, ensuring that each battery stack is charged evenly while avoiding overcharging or undercharging.

[0026] Communication module: Enables communication and information exchange between modules, ensuring coordinated system operation. This module employs a high-speed communication protocol to guarantee the real-time performance and reliability of data transmission.

[0027] Electrolyte concentration distribution affects the performance and lifespan of the fuel cell stack. To further optimize the charging process and ensure charging safety and balance, this system also includes a charging current adjustment module, which adjusts the charging current distribution scheme according to the electrolyte concentration difference inside the fuel cell stack. This allows for a more uniform electrolyte concentration distribution, thereby optimizing the charging process and extending the lifespan of the fuel cell stack.

[0028] The system described in this embodiment also includes a reset module, which is used to perform a reset operation after charging is completed. The reset operation includes resetting the APSO parameters, restoring the position and velocity of the particle swarm to the initial state, resetting the health status indicators of the health status assessment module, restoring the health status indicators of each stack to the initial value, and resetting the charging current adjustment parameters of the charging control module, adjusting the output current of the charging power supply to the initial value.

[0029] Example 2 This embodiment discloses a method for equalizing charging a multi-pile parallel system. This method is based on the system described in Embodiment 1. Figure 2 As shown, the process includes six steps: data acquisition and preprocessing, health status assessment, APSO algorithm optimization, charging control execution, real-time monitoring and adjustment, and charging termination and system reset. The following is a detailed introduction to each step.

[0030] Step 1: Data acquisition and preprocessing.

[0031] Data Acquisition: The data acquisition module collects electrical parameters such as voltage, current, and temperature of each fuel cell stack in real time, as well as health status parameters such as the number of cycles, capacity decay rate, and internal resistance changes of the fuel cell stack.

[0032] In this implementation, it is assumed that there are 3 fuel cells in the system, and the collected data is shown in the table below: Data preprocessing: The collected data is preprocessed, including filtering and normalization, to eliminate noise interference and differences in data units. A moving average filtering algorithm can be used to filter the voltage and current data, normalizing them to the [0,1] interval.

[0033] The normalized data are shown in the table below: The voltages of the three fuel cells are normalized here (lateral normalization of different fuel cells at the same time).

[0034] Step 2: Health status assessment, the process is as follows Figure 3 As shown.

[0035] Health Status Indicator Calculation: The health status assessment module calculates the health status index for each fuel cell stack based on the preprocessed data. The health status index H is calculated using the following formula: , in , , This is a weighting coefficient, which can be adjusted according to the actual characteristics of the fuel cell stack. In specific implementations, it can be set... =0.4、 =0.3、 =0.3, then the health status index of each fuel cell stack is calculated as follows: The calculated health status index is used as a constraint condition for the APSO algorithm to optimize the charging current allocation scheme.

[0036] Step 3: Initialize the system and parameters.

[0037] Startup System: Start the flow stack equalization charging control system and initialize the working status of each module.

[0038] Initialize APSO algorithm parameters: The particle swarm size can be set to 30, the maximum number of iterations to 100, the initial inertia weight to 0.9, and the learning factors c1 and c2 to 2.0.

[0039] Initialize the particle swarm: Assume the particle swarm size is 30, and the position X of each particle is... i It is a vector of length 3 (number of fuel cells), with the initial position randomly generated and the velocity initialized to zero.

[0040] For example, the positions and velocities of the initial particle swarm are shown in the table below: Step 4: APSO algorithm optimizes charging current distribution, the process is as follows: Figure 4 As shown.

[0041] Calculate fitness value: fitness function The fitness function is defined as a comprehensive evaluation based on factors such as the voltage, current, and health status indicators of the fuel cell stack. , in Let represent the fitness value of the i-th particle, and n represent the number of fuel cells. , Let represent the voltage and current of fuel cell j, respectively. , The average voltage and average current of all fuel cells are given. For the health state of the j-th fuel cell stack, This represents the weighting coefficient for the health status of the fuel cell stack.

[0042] Assumption =0.1, then the fitness value of each particle is calculated as follows: If the current particle's fitness value is better than the fitness value of its individual best position, then the individual best position is updated. The individual best position is the best position each particle has ever experienced. If the fitness value corresponding to the current particle's position is better than the fitness value of the previously recorded individual best position, then this new position is updated to the individual best position. If the current particle's fitness value is better than the fitness value of the global best position, then the global best position is updated to the position of this particle. The global best position is the best position experienced by all particles. If the fitness value of a particle is better than the fitness value of the current global best position, then the global best position is updated to the position of this particle.

[0043] For example, suppose the initial global optimum position is [0.3, 0.3, 0.4] and the fitness value is 0.05. During the iteration process, if it is found that the position of particle 30 is [0.4, 0.3, 0.3] and the fitness value is 0.04, which is better than the fitness value of the global optimum position, then the global optimum position is updated to [0.4, 0.3, 0.3].

[0044] In each iteration, the particle's velocity and position are adjusted according to the update formula of the APSO algorithm. The velocity update formula is: , The formula for velocity update during iteration is: , in Let i be the updated velocity of particle i. Let i be the velocity of particle i before the update. For inertial weights, , As a learning factor, , A random number within the interval [0,1]. For the optimal position of an individual, To be the globally optimal position This represents the position of particle i before the update. Let i be the updated position of particle i. Let be the updated velocity of particle i.

[0045] Current particle position For X i =[0.3,0.3,0.4], speed For V i =[0.1,0.1,0.1], the optimal position of an individual. The optimal position is [0.2, 0.3, 0.5]. Given the values ​​[0.4, 0.3, 0.3], inertia weight w = 0.9, learning factors c1 = 2.0, c2 = 2.0, and random numbers r1 = 0.5, r2 = 0.6, the updated velocity and position are calculated as follows: , .

[0046] Determine if the termination condition is met: If the maximum number of iterations is reached or the fitness value of the globally optimal position meets the preset convergence threshold, then stop the iteration and output the optimal charging current allocation scheme; otherwise, continue iterating. In specific implementations, the maximum number of iterations can be set to 100, and the convergence threshold to 10. -6 .

[0047] Step 5: Execution of charging control and optimization of electrolyte concentration distribution.

[0048] After optimizing the charging current allocation scheme using the APSO algorithm, the charging control execution phase begins. This phase not only performs charging control based on the optimized charging current allocation scheme but also introduces a dynamic charging current adjustment method based on the electrolyte concentration distribution within the fuel cell stack to further optimize the charging process and ensure charging safety and balance. The specific implementation steps are as follows: (1). Adjust the charging current. According to the optimized charging current distribution scheme of the APSO module, the charging control module adjusts the charging current of each stack.

[0049] For example, assuming the optimized charging current allocation scheme is [0.4, 0.3, 0.3], and the total charging current is 100A, then the charging currents of each stack are 40A, 30A, and 30A, respectively.

[0050] (2) Electrolyte concentration distribution monitoring: Multiple electrolyte concentration sensors are installed inside each battery stack to monitor the concentration distribution of the electrolyte at different locations within the stack in real time. These concentration sensors can be distributed at the top, middle, and bottom of the stack to obtain more comprehensive concentration information. The concentration sensors transmit real-time data to the data acquisition and processing module, which, after filtering and normalization, provides accurate concentration data to the charging control module.

[0051] (3) Dynamic charging current adjustment mechanism: Based on the electrolyte concentration distribution, the charging control module dynamically adjusts the charging current. The specific adjustment logic is as follows: Concentration difference assessment: Calculate the electrolyte concentration difference between the upper and lower parts of each fuel cell stack. For example, for fuel cell stack 1, the upper concentration is... The lower concentration is The concentration difference is then... .

[0052] The charging current is adjusted based on the concentration difference. The adjustment formula is: , in The adjusted charging current for fuel cell stack 1. The charging current of fuel cell stack 1 before adjustment. To adjust the coefficient, The electrolyte concentration of fuel cell stack 1 is... Adjustments should be made based on the actual characteristics of the fuel cell stack.

[0053] For example, the assumption =0.05, for fuel cell stack 1, if The adjusted charging current is (rounded to two decimal places): .

[0054] Current distribution adjustment: Based on the adjusted charging current, the charging current of each battery stack is redistributed. The distribution method is as follows: First, calculate the difference between the adjusted total charging current and the original charging current, and then compensate the difference to each charging battery stack according to the optimal charging current distribution scheme calculated in step S03.

[0055] For example, if the charging current of fuel cell stack 1 is adjusted from 40A to 39.99A, the total current will become 99.99A. To maintain a total current of 100A, the charging control module will reallocate the charging current of each fuel cell stack according to the optimal allocation scheme. Initially, the charging current allocation ratio of the three fuel cell stacks is 40%, 30%, and 30%, with a total current of 100A. Therefore, the charging currents of fuel cell stacks 1, 2, and 3 are 40A, 30A, and 30A respectively. Now, due to electrolyte concentration issues, the charging current of fuel cell stack 1 has been adjusted from 40A to 39.99A. This results in a total current of 99.99A, a decrease of 0.01A. To readjust the total current back to 100A, the charging control module will reallocate this 0.01A according to the original allocation ratio of each fuel cell stack. Specifically, it will allocate this 0.01A according to a 4:3:3 ratio. This will increase the current of fuel cell stack 1 by 0.004A, and increase the current of fuel cell stacks 2 and 3 by 0.003A each. After adjustment, the charging current of fuel cell stack 1 became 39.994A, while the charging currents of fuel cell stacks 2 and 3 were 30.003A each. This brought the total current back to 100A.

[0056] This process is not simply about adding or removing current to a particular fuel cell stack, but rather about redistributing the missing current to each stack according to the original proportions, ensuring that the total current remains constant.

[0057] Electrolyte concentration distribution affects the performance and lifespan of the fuel cell stack. By adjusting the charging current based on the concentration difference, a more uniform electrolyte concentration distribution can be achieved, thereby optimizing the charging process and extending the stack's lifespan. Although the proportion of charging current for each stack may vary, the total current remains constant, ensuring both charging safety and improved charging uniformity.

[0058] (4) Perform charging control. The charging control module adjusts the output current of the charging power supply according to the adjusted charging current distribution scheme to achieve precise charging control of each battery stack. PWM (Pulse Width Modulation) technology can be used to adjust the output current of the charging power supply to ensure that the charging current of each battery stack is charged according to the optimized distribution scheme.

[0059] By introducing a dynamic charging current adjustment mechanism based on the electrolyte concentration distribution inside the stack during the charging control execution phase, not only is the balanced distribution of charging current ensured, but the impact of uneven electrolyte concentration on stack performance and lifespan is also effectively resolved, significantly improving the performance and reliability of the flow battery stack charging balance control system.

[0060] Step 6: Real-time monitoring and adjustment, the process is as follows: Figure 5 As shown.

[0061] Real-time monitoring: The data acquisition module monitors the electrical parameters of each fuel cell stack in real time, such as voltage, current, and temperature, as well as the health status parameters of the fuel cell stack, and feeds back the monitoring data to the health status assessment module and the APSO module.

[0062] Health Status Index Update: The health status assessment module recalculates the health status indexes of each fuel cell stack based on real-time monitoring data and feeds the updated health status indexes back to the APSO module. For example, assuming that the health status index of fuel cell stack 1 changes during charging, with the capacity decay rate increasing from 0.02% to 0.03%, the health status assessment module will recalculate the health status indexes of that fuel cell stack and feed the updated indexes back to the APSO module.

[0063] Optimize charging current distribution scheme: The APSO module re-optimizes the charging current distribution scheme based on the updated health status indicators and feeds the optimized scheme back to the charging control module to achieve real-time adjustment and optimization of the charging process.

[0064] The APSO module will re-optimize the charging current distribution scheme based on the new health status indicators to ensure the balance and safety of the charging process.

[0065] Step 7: End charging and system reset. When the charging status of all fuel cells reaches the preset charging completion conditions, such as the fuel cell voltage reaching the rated voltage or the charging current dropping to a certain value, the charging control module stops charging and sends a charging completion signal to the system.

[0066] After receiving the charging completion signal, the system resets each module to prepare for the next charging process. For example, it resets the parameters of the APSO module, restoring the position and velocity of the particle swarm to their initial state; it resets the health status indicators of the health status assessment module, restoring the health status indicators of each stack to their initial values; and it resets the charging current adjustment parameters of the charging control module, adjusting the output current of the charging power supply to its initial value.

[0067] In Embodiments 1 and 2, the charging system and method are used in a series-parallel system of multiple flow battery stacks. In other embodiments, they can also be applied to other fields that require series-parallel charging control of multiple battery stacks, such as electric vehicle charging stations and industrial energy storage systems.

[0068] The above description is merely the basic principle and preferred embodiment of the present invention. Improvements and substitutions made by those skilled in the art based on the present invention are within the scope of protection of the present invention.

Claims

1. A method for equalizing charging a multi-stack series-parallel system, characterized in that: Includes the following steps: S01. Data acquisition and preprocessing: Real-time acquisition of voltage, current, temperature, cycle count, capacity decay rate, and internal resistance change rate of each fuel cell stack, and preprocessing of the acquired data; S02. Stack health status assessment: The pre-processed capacity decay rate, internal resistance change rate and cycle number are weighted and summed to calculate the health status of each stack. S03. Optimize charging current distribution based on APSO algorithm, including: S31. Initialize the particle swarm size, maximum number of iterations, initial inertia weight and learning factor of the APSO algorithm, randomly generate the initial position of the particle swarm, and set the initial velocity of the particle swarm to 0. S32. Calculate the fitness value of each particle using the following formula: ， in Let represent the fitness value of the i-th particle, and n represent the number of fuel cells. , Let these represent the voltage and current of fuel cell j, respectively. , The average voltage and average current of all fuel cells are given. For the health state of the j-th fuel cell stack, Weighting coefficients for the health status of the fuel cell stack; S32. Update the optimal position and adjust the particle velocity and position; S33. Determine if the termination condition is met. If not, repeat steps S32 and S33. If it is met, output the optimal solution for charging current allocation. The optimal solution for charging current allocation is the globally optimal position of the particle swarm generated iteratively. S04. Charging control execution: Adjust the charging current of each battery stack according to the optimal scheme of charging current distribution to achieve balanced charging of each battery stack.

2. The equalization charging method for a multi-stack series-parallel system according to claim 1, characterized in that: In step S04, the charging current of the fuel cell stack is adjusted based on the electrolyte concentration distribution inside the stack. The adjustment method is as follows: The electrolyte concentration at different locations inside the fuel cell stack is monitored to obtain the electrolyte concentration difference within the stack. The charging current is then adjusted based on this concentration difference, using the following formula: ， in The adjusted charging current, The charging current before adjustment. To adjust the coefficient, This represents the electrolyte concentration difference in the fuel cell stack.

3. The equalization charging method for a multi-stack series-parallel system according to claim 2, characterized in that: In order to keep the total charging current unchanged, the charging current of each battery stack is redistributed based on the adjusted charging current. The distribution method is as follows: first, the difference between the adjusted total charging current and the original charging current is calculated, and then the difference is compensated to each charging battery stack according to the optimal scheme of charging current distribution calculated in step S03.

4. The equalization charging method for a multi-stack series-parallel system according to claim 2, characterized in that: The electrolyte concentration at different locations inside the fuel cell stack is monitored in real time using an electrolyte concentration sensor.

5. The equalization charging method for a multi-stack series-parallel system according to claim 3, characterized in that: Electrolyte concentration sensors are located at the top, middle, and bottom of the fuel cell stack.

6. The equalization charging method for a multi-stack series-parallel system according to claim 1, characterized in that: Update the optimal position and adjust the particle velocity and position as follows: If the fitness value of the current particle position is better than the fitness value of its individual optimal position, then update the individual optimal position to the current particle position; if the fitness value of a certain particle is better than the fitness value of the global optimal position, then update the global optimal position to the position of this particle. The formula for velocity update during iteration is: ， The formula for velocity update during iteration is: ， in Let i be the updated velocity of particle i. Let i be the velocity of particle i before the update. For inertial weights, , As a learning factor, , A random number within the interval [0,1]. For the optimal position of an individual, To be the globally optimal position This represents the position of particle i before the update. Let i be the updated position of particle i. Let be the velocity of particle i before the update.

7. The equalization charging method for a multi-stack series-parallel system according to claim 1, characterized in that: Then proceed to step S05, real-time monitoring and adjustment, real-time monitoring of the voltage, current, temperature, number of cycles, capacity decay rate, and internal resistance change rate of each fuel cell stack, and re-execute steps S02, S03, and S04 based on the real-time monitoring data.

8. A leveling charging system for a multi-stack series-parallel system, characterized in that: include: A fuel cell stack module includes multiple fuel cell stacks, each of which includes multiple battery cells for storing and releasing electrical energy; The data acquisition and preprocessing module is used to collect the voltage, current, temperature, number of cycles, capacity decay rate, and internal resistance change rate of each fuel cell stack in real time, and to preprocess the collected data. The health status assessment module evaluates the health status of the fuel cell stack based on the number of cycles, capacity decay rate, and internal resistance change rate after preprocessing. The APSO module employs an improved APSO algorithm to optimize the charging current allocation scheme based on the voltage, current, temperature, and health status of the fuel cell stack. This module finds the optimal charging current allocation scheme by simulating the movement of particles in the search space and automatically adjusts the APSO algorithm parameters according to the dynamic changes of the fuel cell stack, thereby improving the convergence speed and global optimization capability of the APSO algorithm. The charging control module controls the charging current of each battery stack according to the charging current distribution scheme.

9. The equalization charging system for a multi-stack series-parallel system according to claim 8, characterized in that: It also includes a charging current adjustment module, which is used to adjust the charging current distribution scheme according to the electrolyte concentration difference inside the stack.

10. The equalization charging system for a multi-stack series-parallel system according to claim 9, characterized in that: It also includes a reset module for performing a reset operation after charging is completed. The reset operation includes resetting the APSO parameters, restoring the position and velocity of the particle swarm to the initial state, and resetting the health status indicators of the health status assessment module, restoring the health status indicators of each stack to the initial value. Reset the charging current adjustment parameters of the charging control module to adjust the output current of the charging power supply to the initial value.

Images