A method for optimizing a proton exchange membrane fuel cell shutdown strategy

Abstract

The application discloses a kind of proton exchange membrane fuel cell shutdown strategy optimization method, belong to proton exchange membrane fuel cell control technical field, by proposing fusion internal current uniformity and external voltage safety dual-index quantitative evaluation framework, based on multidimensional experimental data evaluation obtains basic optimization shutdown strategy;Quasi-one-dimensional dynamic equivalent circuit model of shutdown process is constructed, and optimization algorithm is used to complete model parameter identification and precision improvement;Optimized model is used to re-optimize experimental screening strategy, and the ladder load reduction shutdown strategy with implementation and durability guarantee is proposed, while realizing fast shutdown, maintaining current distribution uniformity, and inhibiting voltage reverse polarity risk.The application is suitable for various power levels proton exchange membrane fuel cell stacks, especially large-area commercial stacks, has wide application prospect, helps to improve the reliability and durability of fuel cell system long-time service, thereby supporting the application and popularization of hydrogen energy in the field of transportation.

Description

Technical Field

[0001] This invention belongs to the field of proton exchange membrane fuel cell control technology, specifically relating to a method for optimizing a proton exchange membrane fuel cell shutdown strategy. Background Technology

[0002] Proton exchange membrane fuel cells (PEMFCs) have become an important development direction for new energy vehicle power systems due to their high efficiency and zero emissions, and have already achieved initial industrial application in the field of hydrogen fuel cell commercial vehicles. However, frequent start-stop operations in actual operation significantly accelerate performance degradation, becoming a key factor restricting the lifespan and large-scale commercial promotion of PEMFCs. If the shutdown strategy is unreasonable, hydrogen-air interfaces can easily form during shutdown, inducing reverse current, which in turn exacerbates catalyst carbon support corrosion and battery performance degradation. At the same time, most existing studies evaluate the shutdown process based on the overall voltage change of the stack, lacking in-depth analysis of the dynamic response characteristics inside the stack, especially the non-uniformity and localized evolution behavior of the internal current distribution. Therefore, existing shutdown strategy evaluation methods cannot accurately reflect the real impact of the shutdown process on stack durability, and there is a lack of unified, quantitative, and engineering-guided shutdown strategy optimization methods.

[0003] Therefore, there is an urgent need to propose a quantitative evaluation method that can comprehensively characterize the uniformity of internal current distribution and the safety of external voltage during PEMFC shutdown, and to establish an optimized shutdown strategy that combines speed, safety and durability based on this evaluation system, so as to improve the long-term service reliability and engineering application stability of fuel cell systems. Summary of the Invention

[0004] To address the shortcomings of existing shutdown strategy evaluation methods, which fail to accurately reflect the true impact of shutdown processes on stack durability and lack a unified, quantitative, and engineering-guided shutdown strategy optimization method, this invention provides a shutdown strategy optimization method for proton exchange membrane fuel cells (PEMFCs). This method provides quantitative basis and a new approach for PEMFC shutdown strategy optimization, which helps improve the reliability and durability of fuel cell systems during long-term operation, thereby supporting the application and promotion of hydrogen energy in the transportation sector.

[0005] The technical solution adopted in this invention is as follows:

[0006] A method for optimizing shutdown strategies of a proton exchange membrane fuel cell includes the following steps:

[0007] Step 1: Under various preset shutdown strategies, collect the overall stack voltage timing data and the partition current density timing data during the shutdown process of the proton exchange membrane fuel cell.

[0008] Step 2: Based on historical shutdown experimental data and safe operating boundaries of proton exchange membrane fuel cells, a dual-index quantitative evaluation framework integrating voltage response characteristics and current distribution uniformity is constructed. The specific process is as follows:

[0009] Step 2.1: Establish voltage safety evaluation indicators, based on downtime t. SD Voltage rebound amplitude ΔV rebound and the lowest voltage of the fuel cell stack V min The scoring functions of the three are obtained by weighted summation;

[0010] Step 2.2: Establish an evaluation index for current distribution uniformity, based on the standard deviation σ of the maximum spatial distribution of current. max The scoring functions for the overall skewness S of the current distribution are obtained by weighted summation;

[0011] Step 2.3: Weighted summation of voltage safety evaluation index and current distribution uniformity evaluation index to obtain the comprehensive score of the dual-index quantitative evaluation framework;

[0012] Step 3: Using a dual-index quantitative evaluation framework, based on the overall voltage time series data and partitioned current density time series data obtained in Step 1, evaluate and rank various preset shutdown strategies, and select the one with the best overall performance as the basic optimized shutdown strategy.

[0013] Step 4: For the shutdown process of the proton exchange membrane fuel cell, the anode hydrogen flow channel of the proton exchange membrane fuel cell is evenly divided into M sub-sections according to the hydrogen flow direction. Each sub-section corresponds to a basic equivalent circuit model. Adjacent sub-sections are connected by a lateral resistor R. t By connecting them in parallel, a basic model of M equivalent circuits and M-1 transverse resistors R are constructed. t The quasi-one-dimensional dynamic equivalent circuit model is constructed; M is a positive integer greater than 1;

[0014] The basic equivalent circuit model includes a cathode loop, a common loop, and an anode loop connected in series; wherein, the cathode loop is composed of an equivalent circuit for the cathode reaction process and an equivalent circuit for the cathode propagation time connected in series, and the equivalent circuit for the cathode propagation time is composed of the cathode parasitic capacitance C. ca and cathode interface transfer resistance R ca The anode circuit is formed by parallel connection; the anode circuit is formed by series connection of the equivalent circuit of the anode reaction process and the equivalent circuit of the anode propagation time, and the equivalent circuit of the anode propagation time is formed by the anode parasitic capacitance C. an and anode interface transfer resistance R an The circuit is formed by parallel connections; the common circuit is formed by the membrane resistor R0 and the controllable voltage source E connected in series.

[0015] Step 5: Using a metaheuristic optimization algorithm, based on the total voltage timing data and partitioned current density timing data obtained in Step 1, the parameters of the quasi-one-dimensional dynamic equivalent circuit model are optimized to obtain the optimized quasi-one-dimensional dynamic equivalent circuit model.

[0016] Step 6: Using the optimized quasi-one-dimensional dynamic equivalent circuit model, simulate and optimize the key control parameters of the basic optimized shutdown strategy, generate multiple sets of candidate shutdown strategies, as well as the corresponding overall voltage timing data and partitioned current density timing data.

[0017] Step 7: Using a dual-index quantitative evaluation framework, based on the overall stack voltage time series data and partition current density time series data generated in Step 6, evaluate and rank multiple candidate shutdown strategies, and select the one with the best overall performance as the deep optimization shutdown strategy.

[0018] Furthermore, in step 1, a high-resolution partition detection device is used to acquire partition current density time-series data.

[0019] Furthermore, the preset shutdown strategy described in step 1 includes a natural shutdown strategy, a gas purging shutdown strategy, and shutdown strategies based on different auxiliary load currents.

[0020] Furthermore, in step 2.1, the downtime t SD Minimum voltage of fuel cell stack V min Voltage rebound amplitude ΔV rebound The weights of the scoring function decrease sequentially, and the voltage rebound amplitude ΔV rebound The weight of the scoring function shall not be less than 15%.

[0021] Furthermore, in step 2.2, the standard deviation σ of the maximum spatial distribution of the current max The weight of the scoring function is greater than the weight of the scoring function of the overall skewness S of the current distribution, and the weight of the scoring function of the overall skewness S of the current distribution is not less than 30%.

[0022] Furthermore, in step 2.3, the weight of the voltage safety evaluation index is greater than the weight of the current distribution uniformity evaluation index, and the weight of the current distribution uniformity evaluation index is not less than 40%.

[0023] Furthermore, the equivalent circuit of the cathode reaction process in step 4 consists of the double-layer capacitance C at the cathode electrode-electrolyte interface. tc Cathode charge transfer resistance R tc and cathode leakage resistance R lc The circuit is formed by parallel connections, and the equivalent circuit for the anolyte reaction process consists of the double-layer capacitance C at the anolyte-electrolyte interface. ta Anode charge transfer resistance R ta and anode leakage resistance R la They are connected in parallel.

[0024] Further, the metaheuristic optimization algorithm described in step 5 is an evolutionary algorithm, a swarm intelligence algorithm, a physics-inspired algorithm, or a hybrid improved algorithm; wherein, the evolutionary algorithm is selected from any one of genetic algorithm, differential evolution, adaptive differential evolution, success history adaptive differential evolution, linear population size reduction adaptive differential evolution, and two-stage differential evolution; the swarm intelligence algorithm is selected from any one of particle swarm optimization, gray wolf optimization, whale optimization, artificial bee colony, tunicate swarm, sparrow search, manta ray foraging, dung beetle optimization, seahorse optimization, and cougar optimization; the physics-inspired algorithm is selected from any one of simulated annealing, gravity search, multi-strategy enhanced snow melting optimization, and enzyme action optimization; the hybrid improved algorithm is selected from any one of particle swarm-genetic hybrid algorithm, differential evolution-particle swarm hybrid algorithm, gray wolf-cuckoo hybrid algorithm, and sparrow search-differential evolution hybrid algorithm.

[0025] Furthermore, in step 6, when the basic optimized shutdown strategy is a natural shutdown strategy, its key control parameter is the gas flow rate; when the basic optimized shutdown strategy is a gas purging shutdown strategy, its key control parameters are the gas flow rate and purging time; when the basic optimized shutdown strategy is a shutdown strategy with different auxiliary load currents, its key control parameters are the gas flow rate and load application method.

[0026] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0027] 1. This invention proposes an optimization method for shutdown strategy of proton exchange membrane fuel cells. It proposes a dual-index quantitative evaluation framework integrating internal current uniformity and external voltage safety, and obtains a basic optimized shutdown strategy based on multi-dimensional experimental data. Secondly, it constructs a quasi-one-dimensional dynamic equivalent circuit model of the shutdown process, and uses an optimization algorithm to identify model parameters and improve accuracy. Subsequently, it uses the optimized model to further optimize the experimental screening strategy, proposing a stepped load reduction shutdown strategy that combines feasibility and durability. This strategy can achieve rapid shutdown while maintaining current distribution uniformity and effectively suppressing the risk of voltage reversal.

[0028] 2. The dual-index quantitative evaluation framework proposed in this invention breaks through the limitations of traditional evaluation that relies solely on external voltage. It can comprehensively and accurately assess the impact of the shutdown process on the durability of the fuel cell stack, providing a scientific quantitative basis for the optimization of shutdown strategies.

[0029] 3. This invention combines high-resolution partitioned experimental detection with quasi-one-dimensional dynamic equivalent circuit model simulation, realizing an optimization process of first screening by experiments and then optimizing the model. This not only ensures the reliability of the optimization results, but also greatly improves the efficiency of strategy optimization and reduces experimental costs.

[0030] 4. This invention is applicable to proton exchange membrane fuel cell stacks of various power levels, especially large-area commercial stacks, and has broad application prospects. It helps to improve the reliability and durability of fuel cell systems during long-term service, thereby supporting the application and promotion of hydrogen energy in the transportation field. Attached Figure Description

[0031] Figure 1 This is an overall flowchart of the proton exchange membrane fuel cell shutdown strategy optimization method proposed in Example 1;

[0032] Figure 2 This is a structural diagram of the dual-index quantitative evaluation framework in Example 1;

[0033] Figure 3 This is a schematic diagram of the basic model of the equivalent circuit in Example 1;

[0034] Figure 4 This is a schematic diagram of the sub-partitioning and modeling of the equivalent circuit model of the fuel cell in Example 1;

[0035] Figure 5 This is a schematic diagram of the proton exchange membrane fuel cell stack in Example 1;

[0036] Figure 6 The simulation results of the shutdown process of the quasi-one-dimensional dynamic equivalent circuit model in Example 1 under the condition of maintaining the hydrogen flow rate on the anode side at 15 SLPM and shutting off the air flow rate on the cathode side when an auxiliary load current of 10 A is applied.

[0037] Figure 7 The simulation results of the shutdown process of the quasi-one-dimensional dynamic equivalent circuit model in Example 1 when a 10A auxiliary load current is applied under the condition of maintaining an air flow rate of 15SLPM on the cathode side and shutting off the hydrogen flow rate on the anode side;

[0038] Figure 8 The total score for multiple preset shutdown strategies in Example 1;

[0039] Figure 9 The current density simulation error curve of the optimized quasi-one-dimensional dynamic equivalent circuit model in Example 1;

[0040] Figure 10 The total score of the multiple candidate shutdown strategies in Example 1;

[0041] The diagram is labeled as follows: A1 - Anode end plate insulation plate, A2 - Anode perforated current collector plate, A3 - High-resolution zoned current acquisition plate, A4 - Anode end graphite plate, A5 - Membrane electrode, A6 - Multi-cell series battery, A7 - Cathode end graphite plate, A8 - Cathode current collector plate, A9 - Cathode end insulation plate, B1 - Hydrogen inlet, B2 - Cooling water outlet, B3 - Air outlet, B4 - Schematic diagram of hydrogen flow path, B5 - Air inlet, B6 - Cooling water inlet, B7 - Hydrogen outlet. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0043] Example 1

[0044] This embodiment proposes an optimization method for proton exchange membrane fuel cell shutdown strategy, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0045] Step 1: A schematic diagram of the large-area commercial proton exchange membrane fuel cell stack structure used in this embodiment is shown below. Figure 5 As shown, it includes, in sequence, an anode end plate insulating plate A1, an anode hollow current collector plate A2, a high-resolution partitioned current acquisition plate A3, an anode end graphite plate A4, a membrane electrode A5, a multi-cell battery connected in series A6, a cathode end graphite plate A7, a cathode current collector plate A8, and a cathode end insulating plate A9.

[0046] Based on the independently developed high spatiotemporal resolution partition detection device A3, the internal current distribution of the proton exchange membrane fuel cell stack is detected in real time during the shutdown process. The detection device can realize the synchronous acquisition of current signals of multiple partitions inside the stack and record the dynamic response characteristics inside the stack during the shutdown process with millisecond-level time resolution.

[0047] The experiment employed multiple preset shutdown strategies, specifically natural shutdown, gas purging shutdown, and shutdown strategies with different auxiliary load currents (5 A, 10 A, 20 A, and 30 A) to shut down the fuel cell stack. The average voltage of the stack (i.e., the overall stack voltage) and the changes in current within each zone were recorded simultaneously during the shutdown process, i.e., the overall stack voltage time series data and the zone current density time series data.

[0048] Step 2: Based on historical shutdown experimental data and safe operating boundaries of proton exchange membrane fuel cells, construct a dual-index quantitative evaluation system integrating voltage safety and current uniformity, such as... Figure 2 As shown, the specific process is as follows:

[0049] Step 2.1: Establish voltage safety evaluation indicators, based on downtime t. SD Voltage rebound amplitude ΔV rebound and the lowest voltage of the fuel cell stack V min The scoring functions of the three are obtained by weighted summation; where the downtime t SD Used to measure shutdown speed and duration of high-potential exposure; voltage rebound amplitude ΔV rebound Used to characterize the rebound of the stack voltage and its electrochemical instability after the load is disconnected; minimum stack voltage V min Used to quantify the risk of anti-polarization and the extent of local damage it may cause.

[0050] In this embodiment, the downtime scoring function Voltage rebound amplitude scoring function And the lowest voltage scoring function of the fuel cell stack The formulas are as follows:

[0051]

[0052]

[0053]

[0054] When performing weighted summation, the downtime t SD The weight is set to the highest level to highlight the priority of rapid shutdown in engineering applications, while also taking into account voltage safety margin, thus affecting the scoring of voltage safety evaluation indicators. for:

[0055]

[0056] Step 2.2: Establish an evaluation index for current distribution uniformity, based on the standard deviation σ of the maximum spatial distribution of current. max The scoring functions for the current distribution and the overall skewness S are obtained by weighted summation; where σ is the maximum spatial standard deviation of the current distribution. max The global non-uniformity of current distribution is used to quantify this; a larger value indicates poorer consistency in the overall electrochemical reaction and a higher risk of overall degradation. The overall skewness S of the current distribution is a dimensionless index; its positive and negative values ​​characterize the direction of current distribution deviation. The absolute value quantifies the degree of local asymmetry and can accurately detect local extreme anomalies such as reverse current at the flow channel outlet and current accumulation at the inlet. A larger absolute value indicates a more severe local electrochemical reaction imbalance and a higher risk of local degradation. From a mechanistic perspective, current distribution non-uniformity is an external manifestation of the imbalance in the internal electrochemical reaction of the battery and is directly related to durability damage such as carbon corrosion and membrane electrode aging. σ max It forms a strong complementary relationship with S, which can make up for the limitations of evaluation by a single indicator, namely σ.max At lower levels, a larger absolute value of S can still identify hidden local high attenuation risks, enabling a comprehensive and accurate characterization of the current distribution characteristics and potential durability risks during shutdown.

[0057] In this embodiment, the scoring function for the standard deviation of the maximum spatial distribution of current is used. and the overall skewness scoring function of current distribution The formulas are as follows:

[0058]

[0059]

[0060] When performing a weighted summation, σ max The scoring quantifies the global spatial uniformity of the current distribution within a battery, which is a core fundamental indicator reflecting the overall consistency of the electrochemical reaction. Therefore, it is given a higher weight, thus increasing the scoring of the current distribution uniformity evaluation index. for:

[0061]

[0062] Step 2.3: Weighted summation of the voltage safety evaluation index and the current distribution uniformity evaluation index to obtain the comprehensive score of the dual-index quantitative evaluation framework. :

[0063]

[0064] This achieves a balance between the speed and safety of shutdown, providing a unified, quantifiable, and engineering-guided comparative standard for evaluating the performance of different shutdown strategies.

[0065] Step 3: Using a dual-index quantitative evaluation framework, based on the overall stack voltage time series data and zoned current density time series data obtained in Step 1, evaluate and rank various preset shutdown strategies. The results are as follows: Figure 8 As shown, the optimal overall performance was selected as the basic optimization shutdown strategy, namely: maintaining the anode hydrogen flow rate at 15 SLPM, closing the cathode air inlet and outlet valves, and simultaneously applying an auxiliary load current of 10 A to the fuel cell stack.

[0066] Step 4: Regarding the shutdown process of the proton exchange membrane fuel cell, such as... Figure 4 As shown, the anode hydrogen flow channel B4 of the proton exchange membrane fuel cell is evenly divided into 8 sub-sections according to the hydrogen flow direction. Each sub-section corresponds to a basic equivalent circuit model. Adjacent sub-sections are connected by a lateral resistor R. t By connecting them in parallel, a basic model consisting of 8 equivalent circuits and 7 transverse resistors R is constructed. t The quasi-one-dimensional dynamic equivalent circuit model is constructed. Figure 4 In the diagram, B1 is the hydrogen inlet, B2 is the cooling water outlet, B3 is the air outlet, B5 is the air inlet, B6 is the cooling water inlet, and B7 is the hydrogen outlet.

[0067] The structure of the basic model of the equivalent circuit is as follows: Figure 3 As shown, it includes a cathode circuit, a common circuit, and an anode circuit connected in series; wherein, the cathode circuit is composed of an equivalent circuit of the cathode reaction process and an equivalent circuit of the cathode transport time connected in series, and the equivalent circuit of the cathode reaction process is composed of the double-layer capacitance C at the cathode electrode-electrolyte interface. tc Cathode charge transfer resistance R tc and cathode leakage resistance R lc The cathode is formed by parallel connections, and the cathode transport time equivalent circuit is composed of the cathode parasitic capacitance C. ca and cathode interface transfer resistance R ca The circuit is formed by parallel connections; the anode circuit is formed by series connection of the equivalent circuit of the anode reaction process and the equivalent circuit of the anode transport time. The equivalent circuit of the anode reaction process is formed by the double-layer capacitance C at the anode electrode-electrolyte interface. ta Anode charge transfer resistance R ta and anode leakage resistance R la The circuit is formed by parallel connections, and the equivalent circuit for anode propagation time is composed of the anode parasitic capacitance C. an and anode interface transfer resistance R an They are connected in parallel; the common circuit is formed by the membrane resistor R0 and the controllable voltage source E connected in series.

[0068] In the basic model of the equivalent circuit, the current flows through R... tc R ta The transmission corresponds to the electrochemical reaction kinetics of the oxygen reduction and hydrogen oxidation reactions; the reverse current passes through R. lc R la Transmission, R lc Equivalent characterization of the impedance characteristics of cathode catalyst carbon support corrosion and platinum particle redox during shutdown, R la The equivalent impedance corresponding to the platinum redox reaction in the anolyte catalyst layer; E simulates the battery voltage drop characteristics caused by the consumption of reactive gas during shutdown; R0 is the ohmic impedance of the proton exchange membrane, reflecting the resistance to proton migration and transport within the membrane; R ca R an The interfacial contact resistances of the corresponding electrode catalytic layer and gas diffusion layer (GDL), the GDL and the bipolar plate, and the combined electron and proton conduction resistance inside the porous electrode were characterized equivalently; C ca C anThe geometric capacitance and porous electrode structure capacitance between the corresponding electrode catalyst layer and GDL are characterized respectively; the transmission time constant determined by the cathode transmission time equivalent circuit and the anode transmission time equivalent circuit directly regulates the transient response behavior of the battery potential during shutdown, and can be used to describe the dynamic changes of voltage and current during shutdown.

[0069] E specifically uses the Nernst voltage, and its expression is:

[0070]

[0071] In the formula, For fuel cell operating temperature, For the effective partial pressure of hydrogen, This is the effective partial pressure of oxygen.

[0072] The quasi-one-dimensional dynamic equivalent circuit model was simulated during a shutdown process under the condition of maintaining a hydrogen flow rate of 15 SLPM on the anode side and shutting off the air flow rate on the cathode side while applying a 10 A auxiliary load current. The initial values ​​of the model parameters are shown in Table 1, and the simulation results are as follows. Figure 6 As shown; (a1) simulation results of section voltage; (b1) simulation results of internal current density of sub-partitions 1 and 2; (b2) simulation results of internal current density of sub-partitions 3 and 4; (b3) simulation results of internal current density of sub-partitions 5 and 6; (b4) simulation results of internal current density of sub-partitions 7 and 8.

[0073] Table 1

[0074]

[0075] The quasi-one-dimensional dynamic equivalent circuit model was simulated during a shutdown process under the condition of maintaining a cathode-side air flow rate of 15 SLPM and shutting off the anode-side hydrogen flow rate while applying a 10 A auxiliary load current. The initial values ​​of the model parameters are shown in Table 2, and the simulation results are as follows. Figure 7 As shown; (a1) simulation results of section voltage; (b1) simulation results of internal current density of sub-partitions 1 and 2; (b2) simulation results of internal current density of sub-partitions 3 and 4; (b3) simulation results of internal current density of sub-partitions 5 and 6; (b4) simulation results of internal current density of sub-partitions 7 and 8.

[0076] Table 2

[0077]

[0078] Step 5: Employing a multi-strategy enhanced snow melting optimization algorithm, based on the overall pile voltage timing data and partitioned current density timing data obtained in Step 1, parameter optimization is performed on the quasi-one-dimensional dynamic equivalent circuit model. By minimizing the error between the model output and experimental data, the prediction accuracy of the model is improved, resulting in the optimized quasi-one-dimensional dynamic equivalent circuit model. The model's output error verification results are as follows: Figure 9 As shown.

[0079] Step 6: Using the optimized quasi-one-dimensional dynamic equivalent circuit model, simulate and optimize the gas flow rate and load application method of the basic optimized shutdown strategy, generate three sets of candidate shutdown strategies, as well as the corresponding overall stack voltage timing data and partition current density timing data.

[0080] The three sets of candidate shutdown strategies generated in this embodiment are as follows:

[0081] The first group is a two-stage stepped load reduction scheme, which adopts a high-to-low loading logic. First, it loads with a 10 A load for 5 seconds, and then switches to a 6 A load until shutdown, in order to verify the optimization effect of the two-stage load reduction on shutdown characteristics.

[0082] The second group is a three-stage stepped load reduction scheme, which linearly reduces the load in three stages: first, load with 10 A for 3 seconds, then load with 7 A for 3 seconds, and finally switch to 4 A until shutdown. The key is to optimize the matching degree between the multi-stage load and the gas concentration.

[0083] The third group is a continuous linear load reduction scheme, which makes the load current decrease continuously and linearly over time. The initial current is 10 A, and it decreases linearly to 0 A at a rate of 1 A / s. This is used to compare the performance difference between continuous load reduction and segmented step load reduction.

[0084] Step 7: Using a dual-index quantitative evaluation framework, based on the overall stack voltage time-series data and zoned current density time-series data generated in Step 6, evaluate and rank the three candidate shutdown strategies. The evaluation results are as follows: Figure 10 As shown, the strategy with the best overall performance was selected as the deep optimization shutdown strategy, namely: maintaining the anode hydrogen flow rate at 15 SLPM, closing the cathode air inlet and outlet valves, and simultaneously applying a three-stage step-down shutdown strategy.

[0085] It should be noted that this is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for optimizing shutdown strategies of a proton exchange membrane fuel cell, characterized in that, Includes the following steps: Step 1: Under various preset shutdown strategies, collect the overall stack voltage timing data and the partition current density timing data during the shutdown process of the proton exchange membrane fuel cell. Step 2: Based on the battery's historical shutdown test data and safe operating boundaries, construct a dual-index quantitative evaluation framework. The specific process is as follows: Step 2.1: Establish voltage safety evaluation indicators, based on downtime t. SD Voltage rebound amplitude ΔV rebound and the lowest voltage of the fuel cell stack V min The scoring functions of the three are obtained by weighted summation; Step 2.2: Establish an evaluation index for current distribution uniformity, based on the standard deviation σ of the maximum spatial distribution of current. max The scoring functions for the overall skewness S of the current distribution are obtained by weighted summation; Step 2.3: Weighted summation of voltage safety evaluation index and current distribution uniformity evaluation index to obtain the comprehensive score of the dual-index quantitative evaluation framework; Step 3: Using a dual-index quantitative evaluation framework, based on the overall voltage time series data and partitioned current density time series data obtained in Step 1, evaluate and rank various preset shutdown strategies, and select the one with the best overall performance as the basic optimized shutdown strategy. Step 4: For the battery shutdown process, the anode hydrogen flow channel is evenly divided into M sub-sections according to the hydrogen flow direction. Each sub-section corresponds to a basic equivalent circuit model. Adjacent sub-sections are connected by a transverse resistor R. t By connecting them in parallel, a basic model of M equivalent circuits and M-1 transverse resistors R are constructed. t The quasi-one-dimensional dynamic equivalent circuit model is constructed; M is a positive integer greater than 1; The basic equivalent circuit model includes a cathode loop, a common loop, and an anode loop connected in series; wherein, the cathode loop is composed of an equivalent circuit for the cathode reaction process and an equivalent circuit for the cathode propagation time connected in series, and the equivalent circuit for the cathode propagation time is composed of the cathode parasitic capacitance C. ca and cathode interface transfer resistance R ca The anode circuit is formed by parallel connection; the anode circuit is formed by series connection of the equivalent circuit of the anode reaction process and the equivalent circuit of the anode propagation time, and the equivalent circuit of the anode propagation time is formed by the anode parasitic capacitance C. an and anode interface transfer resistance R an The circuit is formed by parallel connections; the common circuit is formed by the membrane resistor R0 and the controllable voltage source E connected in series. Step 5: Using a metaheuristic optimization algorithm, based on the total voltage timing data and partitioned current density timing data obtained in Step 1, the parameters of the quasi-one-dimensional dynamic equivalent circuit model are optimized to obtain the optimized quasi-one-dimensional dynamic equivalent circuit model. Step 6: Using the optimized quasi-one-dimensional dynamic equivalent circuit model, simulate and optimize the key control parameters of the basic optimized shutdown strategy, generate multiple sets of candidate shutdown strategies, as well as the corresponding overall voltage timing data and partitioned current density timing data. Step 7: Using a dual-index quantitative evaluation framework, based on the overall stack voltage time series data and partition current density time series data generated in Step 6, evaluate and rank multiple candidate shutdown strategies, and select the one with the best overall performance as the deep optimization shutdown strategy.

2. The proton exchange membrane fuel cell shutdown strategy optimization method according to claim 1, characterized in that, Furthermore, in step 1, a high-resolution partition detection device is used to acquire partition current density time-series data.

3. The proton exchange membrane fuel cell shutdown strategy optimization method according to claim 2, characterized in that, The preset shutdown strategies mentioned in step 1 include natural shutdown strategy, gas purging shutdown strategy, and shutdown strategies based on different auxiliary load currents.

4. The proton exchange membrane fuel cell shutdown strategy optimization method according to claim 3, characterized in that, The downtime t in step 2.1 SD Minimum voltage of fuel cell stack V min Voltage rebound amplitude ΔV rebound The weights of the scoring function decrease sequentially, and the voltage rebound amplitude ΔV rebound The weight of the scoring function shall not be less than 15%.

5. The proton exchange membrane fuel cell shutdown strategy optimization method according to claim 4, characterized in that, In step 2.2, the standard deviation σ of the maximum spatial distribution of current max The weight of the scoring function is greater than the weight of the scoring function of the overall skewness S of the current distribution, and the weight of the scoring function of the overall skewness S of the current distribution is not less than 30%.

6. The proton exchange membrane fuel cell shutdown strategy optimization method according to claim 5, characterized in that, In step 2.3, the weight of the voltage safety evaluation index is greater than the weight of the current distribution uniformity evaluation index, and the weight of the current distribution uniformity evaluation index is not less than 40%.

7. The method for optimizing the shutdown strategy of a proton exchange membrane fuel cell according to claim 3, characterized in that, In step 4, the equivalent circuit for the cathode reaction process consists of the double-layer capacitance C at the cathode electrode-electrolyte interface. tc Cathode charge transfer resistance R tc and cathode leakage resistance R lc The circuit is formed by parallel connections, and the equivalent circuit for the anolyte reaction process consists of the double-layer capacitance C at the anolyte-electrolyte interface. ta Anode charge transfer resistance R ta and anode leakage resistance R la They are connected in parallel.

8. The method for optimizing the shutdown strategy of a proton exchange membrane fuel cell according to claim 3, characterized in that, The metaheuristic optimization algorithm described in step 5 is an evolutionary algorithm, a swarm intelligence algorithm, a physics-inspired algorithm, or a hybrid improved algorithm. Specifically, the evolutionary algorithm is selected from any one of genetic algorithms, differential evolution, adaptive differential evolution, success history adaptive differential evolution, linear population size reduction adaptive differential evolution, and two-stage differential evolution; the swarm intelligence algorithm is selected from any one of particle swarm optimization, gray wolf optimization, whale optimization, artificial bee colony, tunicate swarm, sparrow search, manta ray foraging, dung beetle optimization, seahorse optimization, and cougar optimization; the physics-inspired algorithm is selected from any one of simulated annealing, gravity search, multi-strategy enhanced snow melting optimization, and enzyme action optimization; and the hybrid improved algorithm is selected from any one of particle swarm-genetic hybrid algorithm, differential evolution-particle swarm hybrid algorithm, gray wolf-cuckoo hybrid algorithm, and sparrow search-differential evolution hybrid algorithm.

9. The method for optimizing the shutdown strategy of a proton exchange membrane fuel cell according to claim 3, characterized in that, In step 6, when the basic optimized shutdown strategy is a natural shutdown strategy, the key control parameter is the gas flow rate; when the basic optimized shutdown strategy is a gas purging shutdown strategy, the key control parameters are the gas flow rate and purging time; when the basic optimized shutdown strategy is a shutdown strategy with different auxiliary load currents, the key control parameters are the gas flow rate and load application method.

Images