A flow field structure optimization method based on proton exchange membrane fuel cell vibration characteristics and the cell

Abstract

A kind of cell flow field structure optimization method based on the vibration characteristics of proton exchange membrane fuel cell and its cell, method steps include: S1, establishing the three-dimensional finite element model of cell;S2, determine the vibration mode and inherent frequency of fuel cell;S3, load acceleration power spectrum density of random vibration load, obtain the structural response characteristics of fuel cell assembly under three-axis random vibration excitation;S4, each flow field structure pre-optimization scheme design is carried out, and the flow field structure geometric model is established;S5, obtain maximum stress and stress distribution coefficient;S6, obtain final optimization scheme;Cell has the flow field structure obtained by any optimization method described above;The present application improves the width ratio of flow channel and ridge by establishing three-dimensional finite element model and modal superposition method for modal analysis, significantly reduces the stress concentration and deformation of assembly under dynamic load environment, and improves the stability of fuel cell;Through simulation and optimization design, reduce stress distribution coefficient, make the stress distribution of fuel cell more uniform, significantly improve the performance and life of cell.

Description

Technical Field

[0001] This invention relates to the field of proton exchange membrane fuel cell technology, and in particular to a method for optimizing the flow field structure of a proton exchange membrane fuel cell based on its vibration characteristics, and the resulting battery. Background Technology

[0002] Fuel cells are devices that directly convert the chemical energy of fuel into electrical energy through an electrochemical reaction, offering advantages such as low noise, zero pollution, and high efficiency. Among them, proton exchange membrane fuel cells (PEMFCs) are widely used in transportation, aerospace, and portable power sources due to their low operating temperature. A PEMFC mainly consists of a flow field plate, a gas diffusion layer, a catalyst layer, and a proton exchange membrane. The flow field structure within the flow field plate plays a crucial role in the temperature distribution, current density distribution, and water distribution within the cell, significantly affecting its local and overall performance. Therefore, designing the cell's flow field structure can not only improve the overall performance of the PEMFC but also help extend its service life.

[0003] Currently, performance optimization for proton exchange membrane fuel cells (PEMFCs) mainly focuses on conventional flow field structure optimization and biomimetic flow field design. Conventional flow fields include serpentine, parallel, and interdigitated flow field structures. By adding blockages and altering their shapes within conventional flow fields, mass transfer and water management capabilities within the cell can be improved, gas distribution in the flow field can be promoted, thereby improving water distribution, temperature distribution, and electrochemical reaction rates within the cell, enhancing the uniformity of key parameter distribution, and effectively improving cell performance. Biomimetic flow fields utilize biomimetic similarity theory to design biomimetic flow field structures for fuel cells, giving them the characteristics of low fluid flow resistance and uniform distribution found in biological structures. This ensures excellent heat and mass transfer channels within the cell, effectively improving heat dissipation performance and power density. However, fuel cells operate in complex environments. Under environmental loads, internal components experience stress concentration and warping deformation, leading to a significant reduction in porosity, cracks, and delamination, resulting in a sharp decline in cell performance or even failure.

[0004] Chinese patent CN115621484B discloses a method for optimizing the flow field of a polymer membrane fuel cell. This method relies on three-dimensional numerical simulation. By constructing a numerical model of the fuel cell and simulating the gas flow, reaction rate, and heat distribution within the cell based on actual boundary conditions during operation, such as temperature, pressure, and gas concentration, it extracts key simulation parameters affecting fuel cell performance and designs the flow field structure accordingly. For example, embedding pins to form a honeycomb structure can promote uniform gas distribution and improve cell performance. However, this method only analyzes the influence of the flow field structure on the gas reaction rate within the cell, lacking a quantitative analysis of its impact on the stress deformation response within the cell. It fails to link the layout of the fuel cell stack and the resulting stress values ​​to the lifespan of the fuel cell materials.

[0005] In his master's thesis, "Design and Thermal Management of Variable-Diameter Flow Channels for PEM Fuel Cells Based on Enhanced Mass Transfer," Ye Zhijie emphasized the effectiveness of using enhanced mass transfer and pressure difference principles in PEMFC flow channel design. By integrating complex fluid dynamics and geometric changes (such as bosses and diameter variations), the optimized flow channel significantly outperformed traditional designs in terms of power density, reactant distribution uniformity, and operational stability. However, this research focused on a single flow channel design without comprehensively considering the interaction between components within the cell, thus failing to reflect the impact of component stress conditions and stress distribution uniformity on fuel cell performance.

[0006] In their article "Research Status and Development Trends of PEMFC Flow Channel Structures" published in the Journal of Nanjing University of Aeronautics and Astronautics (2021, 53(04):477-503), Liang Fengli, Wen Ranran, et al. summarized the types of bipolar plates and the specific parameters that need to be considered when optimizing the design of bipolar plate flow channel structures. By selecting appropriate and effective bipolar plate geometry designs, problems such as poor water management, uneven reactant distribution, and uneven current distribution can be solved. Modifications to the flow channel design in this study may cause significant changes in fluid dynamic characteristics, including but not limited to the distribution of internal pressure and fluid velocity. The disadvantages of this study are that some flow channel designs are too complex, increasing manufacturing costs and difficulties; high pressure drop problems reduce system efficiency; and insufficient water and thermal management may also lead to performance and lifespan issues. Summary of the Invention

[0007] To overcome the shortcomings of the prior art, the present invention aims to provide a flow field structure optimization method based on the vibration characteristics of proton exchange membrane fuel cells. By using a three-dimensional finite element model and modal superposition method, the stress distribution and distribution coefficient of the random vibration response of the battery module are obtained. The performance of the battery under different dynamic loads is simulated and analyzed in detail. The structural response performance of the battery under different flow field structures is compared and analyzed to achieve optimization of the battery flow field structure. This method can significantly reduce the stress concentration and warping deformation of the battery module and ensure the long-term stable operation of the battery in harsh environmental conditions.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A method for optimizing the flow field structure of a proton exchange membrane fuel cell based on its vibration characteristics includes the following steps:

[0010] Step S1: Establish a three-dimensional finite element model of the battery based on the physical dimensions and structure of the proton exchange membrane fuel cell; the three-dimensional finite element model of the battery is specifically as follows: end plates are respectively provided on both sides, and from the two end plates to the middle, a current collector, a flow field plate, a sealing gasket, a gas diffusion layer (GDL), a catalyst coating membrane (CCM) and a proton exchange membrane (PEM) are respectively provided.

[0011] Step S2: Use the modal superposition method to perform modal analysis on the proton exchange membrane fuel cell using the three-dimensional finite element model obtained in step S1 to determine the mode shapes and natural frequencies of the fuel cell.

[0012] Step S3: Apply the acceleration power spectral density of each mode shape and natural frequency of the proton exchange membrane fuel cell obtained in step S2 to the three-dimensional finite element model of the proton exchange membrane fuel cell to obtain the structural response characteristics of the proton exchange membrane fuel cell assembly under triaxial random vibration excitation.

[0013] Step S4: Based on the structural response characteristics of the proton exchange membrane fuel cell under triaxial random vibration excitation obtained in step S3, the flow channel of the proton exchange membrane fuel cell is divided into n×m regions according to the gas flow direction from the inlet to the outlet and numbered. The flow field structure of each region is pre-optimized and designed. Geometric models of the flow field structure with different flow channels and ridge width ratios are established respectively, where n is greater than or equal to 3 and m is greater than or equal to 3.

[0014] Step S5: Combining the structural response characteristics obtained in Step S3 and the flow field structure geometric models with different flow channel and ridge width ratios established in Step S4, evaluate and calculate the vibration response of key components of proton exchange membrane fuel cells under various flow field structures, and obtain the maximum stress and stress distribution coefficient of the vibration response of key components of proton exchange membrane fuel cells under different flow field structures with different flow channel and ridge width ratios.

[0015] Step S6: Compare the various flow field structure optimization schemes and select the flow field structure with the smallest response stress and stress distribution coefficient of the proton exchange membrane fuel cell module as the final optimization scheme.

[0016] The process of establishing the three-dimensional finite element model in step S1 includes the following steps:

[0017] The key components and their specific dimensions of the proton exchange membrane fuel cell (PEMFC) were determined. Based on this data, the endplate, current collector, flow field plate, gasket, gas diffuser layer (GDL), catalyst coating membrane (CCM), and proton exchange membrane (PEM) of the PEMFC were constructed. The material properties of each key component of the PEMFC, including density, elastic modulus, and Poisson's ratio, were defined. A three-dimensional finite element model of the PEMFC was constructed based on the above key components, their specific dimensions, and material properties, and the mesh was generated using second-order hexahedral elements. The mesh of the catalyst coating membrane (CCM) and gas diffuser layer (GDL) was refined. Boundary conditions were applied, including several sets of bolt-nut fasteners with tightening torques of 5-9 N·m, applied assembly loads, and temperature loads from a constant power heat source applied to the cathode side of the PEM at different ambient temperatures to simulate the redox reaction on the cathode side and reflect thermal stress on the catalyst coating membrane (CCM).

[0018] In step S2, the mode shapes and natural frequencies of the proton exchange membrane fuel cell are determined, and the mass matrix M and stiffness matrix K of the system are generated using finite element analysis software; the eigenvalue problem (K-ω) is then solved. 2 M)Φ=0, thus obtaining the natural frequency ω of the proton exchange membrane fuel cell. i and the corresponding mode shape Φ i The natural frequencies and mode shapes of each order are extracted from the solution.

[0019] In step S3, the acceleration power spectral density of the random vibration load refers to the triaxial random vibration excitation applied to the proton exchange membrane fuel cell. In the random vibration analysis module, the acceleration power spectral density (PSD) functions in the X, Y, and Z axes are defined and then applied as random vibration loads to the relevant parts of the three-dimensional finite element model of the proton exchange membrane fuel cell. This ensures that the acceleration power spectral density (PSD) functions are applied in the X, Y, and Z directions, thereby obtaining the displacement, velocity, and acceleration response of the proton exchange membrane fuel cell, as well as the stress and strain of the structure under different operating conditions.

[0020] In step S4, the pre-optimization scheme design for each flow field structure is as follows: establish a geometric model of the flow field structure with different flow channel and ridge width ratios, and obtain different flow channel widths and ridge widths, as well as different flow channel and ridge width ratios, by changing the width of the flow channel and ridge. The width of the flow channel and ridge is taken to be between 1-2 mm.

[0021] In step S5, the maximum stress and stress distribution coefficient of the vibration response of key components of the proton exchange membrane fuel cell under different flow field structures with different flow channel and ridge width ratios are obtained. Specifically, based on the modal analysis results of the prestress, the mechanical response law of the fuel cell under triaxial random vibration excitation of the Z-axis, Y-axis, and Z-axis is calculated. For transient dynamic analysis, the explicit central difference method is used at 0, ..., t. n Given the time step solution, solve for t. n+1 The solution at the time step rewrites the dynamic equilibrium differential equation as follows:

[0022]

[0023] In the formula, n represents the nth time point; For external forces to form an array, Forming arrays of internal energy;

[0024] The acceleration is integrated over time using the central difference method to obtain the velocity at the midpoint of the current increment step:

[0025]

[0026] In the formula, Indicates time The velocity vector, Indicates time The velocity vector, At time t n The acceleration vector;

[0027] Then, add the displacement at the beginning of the increment step to determine the displacement at the end of the increment step:

[0028]

[0029] In the formula, {u(t n+1 )} represents time t n+1 The displacement vector, {u(t) n )} represents time t n The displacement vector;

[0030] The displacement, velocity, and acceleration at each discrete time point throughout the entire time domain can be calculated using the above integral formula, and the element stress can then be determined using the material constitutive relation.

[0031] By superimposing the element stiffness of the elements within the computational domain and the equivalent nodal loads for varying temperatures, the overall stiffness equation for the temperature load is obtained:

[0032]

[0033] In the formula, The element temperature-equivalent nodal load array [K] is the overall stiffness matrix;

[0034] Solving for the thermal deformation {q} at the nodes of the computational domain yields the result. The thermal deformation at any point within the computational domain can then be calculated using an interpolation function. Substituting this into the element's physical equations reveals:

[0035] {σ} e =[D]({ε} e -{ε} t )=[D][B]q e -[D]·α·ΔT[1 1 1 0 0 0] T

[0036] In the formula, {σ} e Let {ε} denote the effective stress vector, [D] denote the material stiffness matrix, and {ε} e Let {ε} represent the total strain vector. t This represents the thermal strain vector caused by temperature change, where α represents the coefficient of thermal expansion of the material, and ΔT represents the amount of temperature change.

[0037] Thus, the corresponding thermal stress can be determined;

[0038] By observing the frequency domain analysis results, vibration response analysis was performed on each region of each flow field structure design, and the maximum stress and average stress were collected. The stress distribution coefficient was calculated by calculating the ratio of the maximum stress to the average stress.

[0039] Step S6 involves comparing the average stress and stress distribution coefficient of each ridge under three-dimensional random vibration with different ridge width ratios to obtain the optimal flow channel structure.

[0040] A battery having a flow field structure obtained by any of the above-described battery flow field structure optimization methods.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] 1. This invention establishes a three-dimensional finite element model and performs modal analysis using the modal superposition method. Random vibration loads are applied to the model in the Z, Y, and Z directions. By improving the width ratio of the flow channel and the ridge, the stress concentration and deformation of the components are significantly reduced under dynamic load conditions, thereby improving the mechanical stability and reliability of the fuel cell.

[0043] 2. This invention introduces the concept of stress distribution coefficient to characterize the uniformity and stability of the component structure response under dynamic load. Through simulation analysis and optimization design, the stress distribution coefficient is reduced, making the stress distribution of the fuel cell more uniform, thereby significantly improving the performance and service life of the battery.

[0044] In summary, this invention establishes a three-dimensional finite element model and performs modal analysis using the modal superposition method. By introducing the concept of stress distribution coefficient, and through simulation analysis and optimization design, it compares and analyzes the battery structure response performance under different flow field structures, thereby achieving battery flow field structure optimization. This significantly reduces stress concentration and warping deformation of battery components, ensuring long-term stable operation of the battery in harsh environmental conditions. Attached Figure Description

[0045] Figure 1 This is a flowchart of the method of the present invention.

[0046] Figure 2 This is a schematic diagram of the battery structure of the present invention.

[0047] Figure 3 This is a block division diagram of a local area of ​​the battery according to the present invention.

[0048] Figure 4 The deformation distribution of the proton exchange membrane (PEM) under different bolt torques according to the present invention; wherein, Figure 4 (a) is a diagram showing the deformation distribution of the proton exchange membrane (PEM) under bolt torque of 5 N·m. Figure 4 (b) is a diagram showing the deformation distribution of the proton exchange membrane (PEM) under bolt torque of 7 N·m. Figure 4 (c) is a distribution diagram of proton exchange membrane (PEM) deformation under bolt torque of 9 N·m.

[0049] Figure 5 This is a von Mises stress curve diagram of the ridge under different bolt preloads according to the present invention.

[0050] Figure 6 The diagram shows the 3σ average von Mises stress of the gas diffusion layer of the present invention under random vibration at different temperatures.

[0051] Figure 7 This is a diagram showing the stress distribution in a local area under different flow field structures according to the present invention.

[0052] Figure 8 This is a dimensionless stress distribution diagram for different flow channel structures of the present invention. Detailed Implementation

[0053] To more clearly illustrate the technical solution and advantages of the present invention, the present invention will be described in detail below with reference to specific implementation steps. It should be noted that the implementation steps described below are only for explaining the present invention and are not intended to limit the present invention.

[0054] Reference Figure 1 The specific steps of implementing this invention are as follows:

[0055] A method for optimizing the flow field structure of a battery includes the following steps:

[0056] Step S1: Establish a three-dimensional finite element model of the proton exchange membrane fuel cell based on its physical dimensions and structure. The specific steps are as follows:

[0057] Step S1.1: Based on the actual dimensions of the fuel cell, create the basic geometry of each component in the software Unigraphics (UG), defining the basic dimensions and thickness of each component as follows: Figure 2 As shown: The gas diffusion layer (GDL) measures 40mm × 40mm with a thickness of 0.3mm; the flow field plate measures 76mm × 76mm with a thickness of 13mm; the sealing gasket measures 76mm × 76mm with a thickness of 0.362mm; the manifold measures 76mm × 108mm with a thickness of 1.3mm; and the end plate measures 108mm × 108mm with a thickness of 19mm. All components are then assembled: At the start of assembly, the gas diffusion layer is placed on the flow field plate, followed by the sealing gasket to prevent gas leakage; then, the catalyst layer and proton exchange membrane are installed, the manifold is placed on both sides of the flow field plate, and finally, the end plate is covered and the entire structure is secured with bolts and nuts to the specified torque.

[0058] Step S1.2: Import the geometric structure obtained in step S1.1 into the finite element analysis software and determine the material properties of each part of the fuel cell, including density, elastic modulus, Poisson's ratio, thermal conductivity, coefficient of thermal expansion and specific heat capacity.

[0059] Step S1.3: Simplify the model by simplifying the mesh generation and reducing computational costs; remove fillets and chamfers from the model to avoid generating smaller and more complex mesh units, reduce mesh generation time, and reduce computational resource requirements; perform mesh generation on the three-dimensional model, with the global mesh mainly composed of second-order hexahedral element meshes, and refine the mesh of the gas diffusion layer (GDL) to finally obtain multiple model regions with different mesh densities.

[0060] Step S1.4: Apply boundary conditions to the three-dimensional model. The assembly load is the assembly of the fuel cell by 8 sets of bolt-nut fasteners on its end plate. Each set of bolt-nuts has a tightening torque of 7 N·m for the fuel cell. The temperature load refers to the heat of electrochemical reaction at the cathode catalyst layer and the overall operation of the proton exchange membrane fuel cell at room temperature.

[0061] Step S2: Determine the mode shapes and natural frequencies of the fuel cell, and perform modal analysis of the fuel cell using the modal superposition method. The specific steps are as follows:

[0062] Step S2.1: In the ANSYS working environment, select modal analysis as the analysis type, and calculate the natural frequencies and corresponding mode shapes of the system through software.

[0063] Step S2.2: Select the main modes that have the greatest impact on the system response, and use the modal superposition method to perform weighted superposition to calculate the total response. The obtained dynamic response results are post-processed and analyzed in finite element software. The total displacement, stress distribution, strain analysis, safety factor, and frequency response formed by the superposition of each main mode can be viewed in ANSYS's Post-Processing tool.

[0064] Step S3: Combining the mode shapes and natural frequencies of the fuel cell obtained in Step S2, the acceleration power spectral density of the random vibration load is applied to obtain the structural response characteristics of the fuel cell assembly under triaxial random vibration excitation. The specific steps are as follows:

[0065] Step S3.1: Define the acceleration power spectral density (PSD) function in ANSYS and apply it in the Z, Y, and Z directions;

[0066] Define the power spectral density function S of a stationary random process Z(t). x (ω) is the autocorrelation function R x The Fourier transform of (τ) is:

[0067]

[0068] Due to S x (ω) is an even function, which can be used to convert the spectral density in the negative frequency range to the positive frequency range, thus obtaining the one-sided power spectral density G. x (ω), that is:

[0069]

[0070] Step S3.2: Use frequency domain analysis to calculate the structural response of each point of the component within a predefined frequency range, such as displacement, velocity, acceleration, and stress; view the results using post-processing tools and generate response spectrum and stress distribution diagrams.

[0071] Step S4: Design pre-optimization schemes for each flow field structure, and establish geometric models of flow field structures with different flow channels and ridge width ratios. The specific steps are as follows:

[0072] Step S4.1, refer to as follows Figure 3The flow channel of the fuel cell was divided into 3×3 regions according to the gas flow direction from inlet to outlet and numbered. The average stress value of the gas diffusion layer at each region was measured, and the stress differences between different regions were compared to evaluate the reliability of different regions under normal operating conditions. Regions with small and uniform stress distribution have longer lifespans, avoiding local stress concentration and enhancing the overall durability and reliability of the structure. Regions with large and uneven stress distribution have shorter lifespans. Stress concentration accelerates local material fatigue, crack formation and propagation, increases the possibility of early failure, and leads to a decrease in the overall durability and reliability of the structure.

[0073] Step S4.2: Change the width of the flow channel and the ridge to obtain different flow field structures. Adjust the width ratio of the flow channel and the ridge from the original 2:2 (unit: mm) to three new ratios: 1:1, 1:2 and 2:1.

[0074] Step S5: Combining the response spectrum and stress distribution diagram from step (3) with the different flow field structures from step (4), obtain the maximum stress and stress distribution coefficient of the vibration response of the key battery components under different flow field structures. The specific steps are as follows:

[0075] Step S5.1: Based on the modal analysis results of prestress, calculate the mechanical response of the fuel cell under triaxial random vibration excitation (Z-axis, Y-axis, Z-axis). For transient dynamic analysis, use the explicit central difference method at 0, ..., t. n Given the time step solution, solve for t. n+1 The solution at the time step rewrites the dynamic equilibrium differential equation as follows:

[0076]

[0077] In the formula, n represents the nth time point; For external forces to form an array, Forming arrays of internal energy;

[0078] The acceleration is integrated over time using the central difference method to obtain the velocity at the midpoint of the current increment step:

[0079]

[0080] In the formula, Indicates time The velocity vector, Indicates time The velocity vector, At time t n acceleration vector

[0081] Then, add the displacement at the beginning of the increment step to determine the displacement at the end of the increment step:

[0082]

[0083] In the formula, {u(t n+1 )} represents time t n+1 The displacement vector, {u(t) n )} represents time t n The displacement vector;

[0084] The displacement, velocity, and acceleration at each discrete time point throughout the entire time domain can be calculated using the above integral formula, and the element stress can then be determined using the material constitutive relation.

[0085] Step S5.2: After superimposing the element stiffness and temperature-equivalent nodal loads of the elements in the computational domain, the overall stiffness equation is obtained:

[0086]

[0087] In the formula, [K] is the element temperature equivalent nodal load array, and [K] is the global stiffness matrix;

[0088] Solving for the thermal deformation {q} at the nodes of the computational domain yields the result. The thermal deformation at any point within the computational domain can then be calculated using an interpolation function. Substituting this into the element's physical equations reveals:

[0089] {σ} e =[D]({ε} e -{ε} t )=[D][B]q e -[D]·α·ΔT[1 1 1 0 0 0] T

[0090] In the formula, {σ} e Let {ε} denote the effective stress vector, [D] denote the material stiffness matrix, and {ε} e Let {ε} represent the total strain vector. t This represents the thermal strain vector caused by temperature change, where α represents the coefficient of thermal expansion of the material, and ΔT represents the amount of temperature change.

[0091] Thus, the corresponding thermal stress can be determined;

[0092] Step S5.3, as described in step S3, involves observing the frequency domain analysis results and performing vibration response analysis on each region of each flow field structure design, such as... Figure 4 The maximum and average stresses are collected to determine the stress concentration areas on the gas diffusion layer.

[0093] Step S5.4, refer to Figure 5Based on the average stress value and stress distribution obtained in step S5.3, the stress distribution coefficient of the component under different flow field structures is calculated according to the different flow field structure geometric models established in step S4. The formula for calculating the average stress is as follows:

[0094]

[0095] In the formula, This is the average stress, where n is the total number of data points, and σ is the average stress. i It is the stress value of the i-th data point.

[0096] Calculate the average stress in regions 1 to 9 of the gas diffusion layer; determine the maximum stress value in different regions, and calculate the stress distribution coefficient for each region using the following formula:

[0097]

[0098] In the formula, K is the stress distribution coefficient, σ max It is the maximum stress. It is the average stress;

[0099] Step S6: Compare the flow field structures with the smallest response stress and stress distribution coefficient of the battery module among the various flow field structure optimization schemes to obtain the final optimization scheme. By comparing the average stress and stress distribution coefficient of each ridge under three-dimensional random vibration with different flow channel ridge width ratios, the optimal flow channel structure optimization method is obtained.

[0100] A battery having a flow field structure obtained by any of the above-described battery flow field structure optimization methods.

[0101] The above description is only one specific embodiment of the present invention and does not constitute any limitation on the present invention. Obviously, those skilled in the art, after understanding the content and design principles of the present invention, may make various modifications and changes in form and detail based on the principles and structure of the present invention. However, these modifications and changes based on the ideas of the present invention are still within the protection scope of the claims of the present invention.

[0102] Example 1

[0103] In the initial design step, the proton exchange membrane battery is initialized to obtain an initial model. This initial model includes: a gas diffusion layer (GDL), a flow field plate, a sealing gasket, a current collector, end plates, and M6 bolts and nuts. During the initial assembly process, the gas diffusion layer is first placed on the flow field plate, followed by the sealing gasket to prevent gas leakage. Subsequently, the catalyst layer and proton exchange membrane are installed, and the current collectors are placed on both sides of the flow field plate. Finally, the end plates are used to cover the structure, and the entire structure is secured with bolts and nuts to the specified torque.

[0104] In this embodiment, the stress distribution of the gas diffusion layer (GDL) was measured under different bolt torque settings, with particular attention paid to stress changes under the ridges and flow channels. The stress changes in the GDL were observed and recorded by varying the bolt preload from 5 N·m to 9 N·m. It was determined that under all bolt preload settings, the stress in the GDL exhibited a concave shape, meaning the stress was lowest at the center and relatively higher at the edges.

[0105] In Example 1, the study of the preload of the fuel cell bolts aims to understand the stress response of the gas diffusion layer (GDL) in a real-world operating environment, particularly its performance under different mechanical loading conditions, in order to optimize the design and improve the overall reliability and efficiency of the fuel cell. This detailed stress analysis allows for better prediction and prevention of potential structural failures due to stress concentration or uneven distribution, thereby extending the fuel cell's lifespan and improving its performance.

[0106] Figure 4 The deformation distribution diagram of the gas diffusion layer (GDL) under different bolt torques in Example 1 of the present invention shows the stress distribution trend: as the bolt torque increases, the stress in the gas diffusion layer increases overall, especially in the region near the flow channel.

[0107] Figure 5 The von Mises stress curves under the ridge under different bolt preloads in Embodiment 1 of the present invention describe the variation of the von Mises stress of the gas diffusion layer (GDL) with radial position under different bolt torques. Each curve corresponds to different torque settings (5 N·m to 9 N·m). It can be seen that due to the stress concentration phenomenon at the flow channel corner, the stress of the gas diffusion layer under the ridge is the largest at x=0, then decreases sharply and gradually decreases, reaching the minimum at the center of the ridge, and then gradually increases along the X direction. With the increase of bolt preload, the stress at all points of the gas diffusion layer under the ridge increases accordingly, especially at the flow channel corner, where the stress overshoot value increases from 0.58 MPa at 5 N·m to 1.15 MPa at 9 N·m, an increase of approximately 98.28%.

[0108] Example 2

[0109] In the initial design step, when initializing the proton exchange membrane fuel cell, temperature changes have a significant impact on the stress response of the gas diffusion layer. Simulation analysis was run under different temperature conditions (such as -30℃, 20℃, and 70℃) to collect stress, strain, displacement data, and von Mises stress of the gas diffusion layer (GDL) under various temperature and load conditions.

[0110] Figure 6The three-sigma average von-Mises stress of the gas diffusion layer under random vibration at different temperatures in Example 2 of this invention is shown below. Figure 8 As shown. Under random vibration excitation along the X-axis, an increase in temperature slightly reduces the average stress of the gas diffusion layer (GDL). Under random vibration excitation along the Y-axis, the GDL exhibits the highest stress level at room temperature; both the maximum and average stresses decrease with increasing or decreasing temperature. Under random vibration excitation along the Z-axis, both the maximum and average stresses show a negative correlation with temperature, decreasing as temperature rises. Under XYZ triaxial random vibration excitation, temperature significantly affects the stress response of the gas diffusion layer. Comparing temperature conditions of 20℃ and 70℃, the maximum stress decreases from 6.33 MPa to 4.60 MPa, a decrease of 1.73 MPa, while the average stress decreases by 0.73 MPa. Furthermore, the decrease in both maximum and average stresses under this condition is greater than under the other three uniaxial vibration conditions.

[0111] Example 3

[0112] In the initial design phase, the widths of the proton exchange membrane fuel cell flow channels are varied to obtain different flow field structures. The width ratio of the flow channels to the ridges is adjusted from the original 2:2 (Unit: mm) to three new ratios: 1:1, 1:2, and 2:1. Finite element analysis is used to predict the stress response of the gas diffusion layer (GDL) under different flow channel and ridge configurations.

[0113] like Figure 7 In Example 3 of this invention, the stress levels of the gas diffusion layer (GDL) varied under different configurations of local stress distribution within different flow field structures. The 1:1 configuration exhibited a moderate stress level, relative stability, and a balanced stress distribution, suitable for general applications. The 1:2 and 2:1 configurations showed higher stress levels most of the time, indicating that these configurations might generate greater local stress in the GDL, suitable for scenarios requiring high hydrodynamic performance, but potentially requiring additional structural reinforcement. The 2:2 configuration showed the lowest stress level, indicating good structural support and uniform force distribution, and is likely the most suitable configuration for long-term stable operation.

[0114] Figure 8To illustrate the dimensionless stress distribution under different flow channel structures in Example 3 of this invention, the average stress ratio of each ridge position under different flow channel ridge width ratios was extracted to the average stress ratio of the ridge positions in the entire flow field. The average stress ratios under the four flow channel ridge width ratios were divided into three parts: top, middle, and bottom. The average stress ratio of the gas diffusion layer (GDL) at the contact position with the middle ridge is close to 1, indicating that the stress at the middle ridge position is relatively uniform compared to the entire flow field. The average stress ratio at the top ridge position is all above 1.1, while the average stress ratio at the bottom ridge position is approximately in the range of 0.85 to 0.9, indicating that the stress uniformity of the gas diffusion layer (GDL) at the bottom ridge position is better than that at the top. Comparing the stress of the transverse blocks at the top, middle, and bottom, the transverse stress uniformity is worst when the flow channel width is 1 mm and the ridge width is 1 mm. When the flow channel width is 2 mm and the ridge width is 2 mm, the transverse stress uniformity at the top and middle of the gas diffusion layer (GDL) is the best.

[0115] As can be seen from the above embodiments and figures, compared with the prior art, the present invention obtains the stress distribution and distribution coefficient of the random vibration response of the battery module through a three-dimensional finite element model and modal superposition method, simulates and analyzes the performance of the battery under different dynamic loads in detail, compares and analyzes the battery structural response performance under different flow field structures, realizes the optimization of the battery flow field structure, and can better predict and prevent potential structural failures caused by stress concentration or uneven distribution, thereby extending the life of the fuel cell and improving its performance. In Embodiment 1, as the bolt preload increases, the stress at all points of the gas diffusion layer under the ridge increases accordingly, especially at the flow channel corner, the stress overshoot value decreases from 1.15 MPa at 9 N·m to 0.58 MPa at 5 N·m. As the stress overshoot value decreases, the stress in the battery affected by the stress increases. The material lifespan in the affected area is significantly increased, specifically, it may increase by 681.2%, or approximately 7.8 times. In Example 2, under XYZ triaxial random vibration excitation, temperature has a significant impact on the stress response of the gas diffusion layer. Comparing temperature conditions of 20℃ and 70℃, the maximum stress decreases from 6.33MPa to 4.60MPa, a decrease of 1.73MPa, and the average stress decreases by 0.73MPa. As the temperature rises to 70℃, the decrease in the maximum stress value will increase the fatigue life of the material by approximately 92%. In Example 3, when the channel width is 2mm and the ridge width is 2mm, the lateral stress uniformity at the top and middle of the gas diffusion layer (GDL) is the best. The stress difference caused by different channel width and ridge width ratios may lead to a significant increase in lifespan. When the channel ridge width ratio increases from 2:1 to 2:2, the lifespan may increase by approximately 196%. Therefore, this invention has innovative points and advantages in achieving optimized battery flow field structure, significantly reducing stress concentration and warpage deformation of battery components, and ensuring long-term stable operation of the battery under harsh environmental conditions.

Claims

1. A method for optimizing the flow field structure of a proton exchange membrane fuel cell based on its vibration characteristics, characterized in that, Includes the following steps: Step S1: Establish a three-dimensional finite element model of the battery based on the physical dimensions and structure of the proton exchange membrane fuel cell; the three-dimensional finite element model of the battery is specifically as follows: end plates are respectively provided on both sides, and from the two end plates to the middle, a current collector, a flow field plate, a sealing gasket, a gas diffusion layer (GDL), a catalyst coating membrane (CCM) and a proton exchange membrane (PEM) are respectively provided. Step S2: Use the modal superposition method to perform modal analysis on the proton exchange membrane fuel cell using the three-dimensional finite element model obtained in step S1 to determine the mode shapes and natural frequencies of the proton exchange membrane fuel cell. Step S3: Apply the acceleration power spectral density of each mode shape and natural frequency of the proton exchange membrane fuel cell obtained in step S2 to the three-dimensional finite element model of the proton exchange membrane fuel cell to obtain the structural response characteristics of the proton exchange membrane fuel cell assembly under triaxial random vibration excitation. Step S4: Based on the structural response characteristics of the proton exchange membrane fuel cell under triaxial random vibration excitation obtained in step S3, the flow channel of the proton exchange membrane fuel cell is divided into n×m regions according to the gas flow direction from the inlet to the outlet and numbered. The flow field structure of each region is pre-optimized and designed. Geometric models of the flow field structure with different flow channels and ridge width ratios are established respectively, where n is greater than or equal to 3 and m is greater than or equal to 3. Step S5: Combining the structural response characteristics obtained in step S3 and the flow field structure geometric models with different flow channel and ridge width ratios established in step S4, evaluate and calculate the vibration response of key components of proton exchange membrane battery under various flow field structures, and obtain the maximum stress and stress distribution coefficient of the vibration response of key battery components under different flow channel and ridge width ratios. Step S6: Compare the various flow field structure optimization schemes and select the flow field structure with the smallest response stress and stress distribution coefficient of the proton exchange membrane battery module as the final optimization scheme.

2. The battery flow field structure optimization method according to claim 1, characterized in that, The process of establishing the three-dimensional finite element model in step S1 includes the following steps: The key components and their specific dimensions of the proton exchange membrane fuel cell (PEMFC) were determined. Based on this data, the endplate, current collector, flow field plate, gasket, gas diffuser layer (GDL), catalyst coating membrane (CCM), and proton exchange membrane (PEM) of the PEMFC were constructed. The material properties of each key component of the PEMFC, including density, elastic modulus, and Poisson's ratio, were defined. A three-dimensional finite element model of the PEMFC was constructed based on the above key components, their specific dimensions, and material properties, and the mesh was generated using second-order hexahedral elements. The mesh of the catalyst coating membrane (CCM) and gas diffuser layer (GDL) was refined. Boundary conditions were applied, including several sets of bolt-nut fasteners with tightening torques of 5-9 N·m, applied assembly loads, and temperature loads from a constant power heat source applied to the cathode side of the PEM at different ambient temperatures to simulate the redox reaction on the cathode side and reflect thermal stress on the catalyst coating membrane (CCM).

3. The battery flow field structure optimization method according to claim 1, characterized in that, In step S2, the mode shapes and natural frequencies of the proton exchange membrane fuel cell are determined, and the mass matrix M and stiffness matrix K of the system are generated using finite element analysis software; the eigenvalue problem (K-ω) is then solved. 2 M)Φ=0, thus obtaining the natural frequency ω of the proton exchange membrane fuel cell. i and the corresponding mode shape Φ i The natural frequencies and mode shapes of each order are extracted from the solution.

4. The battery flow field structure optimization method according to claim 1, characterized in that, In step S3, the acceleration power spectral density of the random vibration load refers to the application of triaxial random vibration excitation to the fuel cell. In the random vibration analysis module, the acceleration power spectral density (PSD) function in the X, Y, and Z axes is defined and then applied as a random vibration load to the relevant parts of the three-dimensional finite element model of the proton exchange membrane fuel cell. This ensures that the acceleration power spectral density (PSD) function is applied in all three directions (X, Y, and Z), thereby obtaining the displacement, velocity, and acceleration response of the proton exchange membrane fuel cell, as well as the stress and strain of the structure under different operating conditions.

5. The battery flow field structure optimization method according to claim 1, characterized in that, In step S4, the pre-optimization scheme design for each flow field structure is as follows: establish a geometric model of the flow field structure with different flow channel and ridge width ratios, and obtain different flow channel widths and ridge widths, as well as different flow channel and ridge width ratios, by changing the width of the flow channel and ridge. The width of the flow channel and ridge is taken to be between 1-2 mm.

6. The battery flow field structure optimization method according to claim 1, characterized in that, In step S5, the maximum stress and stress distribution coefficient of the vibration response of key components of the proton exchange membrane fuel cell under different flow field structures with different flow channel and ridge width ratios are obtained. Specifically, based on the modal analysis results of prestress, the mechanical response law of the proton exchange membrane fuel cell under triaxial random vibration excitation of the X-axis, Y-axis, and Z-axis is calculated. For transient dynamic analysis, the explicit central difference method is used at 0, ..., t. n Given the time step solution, solve for t. n+1 The solution at the time step rewrites the dynamic equilibrium differential equation as follows: In the formula, n represents the nth time point; For external forces to form an array, Forming arrays of internal energy; The acceleration is integrated over time using the central difference method to obtain the velocity at the midpoint of the current increment step: In the formula, Indicates time The velocity vector, Indicates time The velocity vector, At time t n The acceleration vector; Then, add the displacement at the beginning of the increment step to determine the displacement at the end of the increment step: In the formula, {u(t n+1 )} represents time t n+1 The displacement vector, {u(t) n )} represents time t n The displacement vector; The displacement, velocity, and acceleration at each discrete time point throughout the entire time domain can be calculated using the above integral formula, and the element stress can then be determined using the material constitutive relation. By superimposing the element stiffness of the elements within the computational domain and the equivalent nodal loads for varying temperatures, the overall stiffness equation for the temperature load is obtained: In the formula, K is the matrix of the total stiffness, where K is the equivalent nodal load array for the element under varying temperatures. Solving for the thermal deformation q at the nodes of the computational domain yields the thermal deformation at any point within the computational domain. This thermal deformation can then be calculated using an interpolation function. Substituting this into the element's physical equations: {s} e =[D]({e} e -{e} t )=[D][B]q e -[D]·α·ΔT[1 1 1 0 0 0] T In the formula, {σ} e Let {ε} denote the effective stress vector, [D] denote the material stiffness matrix, and {ε} e Let {ε} represent the total strain vector. t This represents the thermal strain vector caused by temperature change, where α represents the coefficient of thermal expansion of the material, and ΔT represents the amount of temperature change. Thus, the corresponding thermal stress can be determined; By observing the frequency domain analysis results, vibration response analysis was performed on each region of each flow field structure design, and the maximum stress and average stress were collected. The stress distribution coefficient was calculated by calculating the ratio of the maximum stress to the average stress.

7. The battery flow field structure optimization method according to claim 1, characterized in that, Step S6 involves comparing the average stress and stress distribution coefficient of each ridge under three-dimensional random vibration with different ridge width ratios to obtain the optimal flow channel structure.

8. A battery, characterized in that, The battery has a flow field structure obtained by a battery flow field structure optimization method as described in any one of claims 1-7.

Images