Fuel cell cathode catalyst layer ionomer gradient robust optimization method and fuel cell

Abstract

The present application belongs to the field of fuel cell catalyst layer design, and particularly relates to a fuel cell cathode catalyst layer ionomer gradient robust optimization method and a fuel cell. The method uses an asymmetric two-section power law curve to represent the ionomer content gradient distribution in the cathode catalyst layer, and has the ability to flexibly control the transition balance of the proton inlet and oxygen inlet side functional area, overall transport support, and local enrichment or reduction; after determining the boundary humidity condition and robust optimization target, a multi-humidity response data set reflecting the mass transfer bottleneck under extreme conditions is generated, and the optimal ionomer gradient scheme is obtained through multi-objective robust optimization and verified within the humidity condition range. The fuel cell with the ionomer gradient distribution determined by the method can alleviate the dominant mass transfer bottleneck under the boundary humidity, and maintain sufficient proton and oxygen transmission, limit abnormal reaction areas under intermediate humidity, thereby realizing stable net output power improvement and limiting the deterioration of reaction uniformity in the catalyst layer within a wide humidity range.

Description

Technical Field

[0001] This invention relates to the field of fuel cell catalyst layer design, and more specifically, to a robust optimization method for ionomer gradient of fuel cell cathode catalyst layer and a fuel cell. Background Technology

[0002] As an electrochemical energy conversion device with broad application prospects, the large-scale application of proton exchange membrane fuel cells (PEMFCs) is limited by factors such as performance, durability, and cost. These issues are closely related to the cathode catalyst layer in the membrane electrode assembly (MEA), and its rational design is of great significance for improving fuel cell performance, durability, and reducing costs. Ionomers in the catalyst layer are key components affecting the cell reaction and transport processes because they provide channels for proton conduction; however, excessively high ionomer content increases oxygen transport resistance. Especially under low platinum loading, the reduction of catalytic active sites exacerbates local transport limitations and reaction inhomogeneity. Therefore, a reasonable balance needs to be achieved between the ionomer content and its spatial distribution and proton conduction, oxygen transport, and catalytic reaction.

[0003] To address these issues, gradient design of ionomer content has gradually gained attention. By adjusting the ionomer content, a high-content region can achieve stronger proton conduction capacity, while a low-content region can achieve better oxygen transport capacity, thereby improving the overall output performance and reaction distribution of the fuel cell. However, existing ionomer gradient designs mostly adopt linear or step-like gradient forms, which have limited degrees of freedom in adjustment, restricting their applicability and robustness under actual operating conditions.

[0004] For example, fuel cells often face varying humidity conditions during actual operation. Relative humidity significantly affects membrane hydration, oxygen dilution, and the risk of liquid water blockage. Therefore, under low humidity conditions, the cell is more susceptible to proton conduction limitations, while under high humidity conditions, oxygen transport limitations are more pronounced. Consequently, the required ionomer distribution differs under different humidity conditions, and current gradient designs obtained for single humidity conditions struggle to maintain good performance and uniformity across a wide humidity range. Existing patent application CN118918979A discloses a method for predicting the volume fraction and gradient distribution of ionomers in a cathode catalyst layer, based on an agglomerate model to obtain the maximum power density under different parameters. A neural network is trained to obtain the optimal ionomer volume fraction for the first gradient-free catalyst layer, and a second optimal value is obtained through a bisection iteration. The gradient is set based on the first optimal value, and the neural network is trained using the ionomer gradient to obtain the first optimal ionomer gradient distribution. The correlation between the ionomer gradient and the average volume fraction is then combined with the second optimal value to obtain the second optimal gradient distribution. This approach focuses on improving the prediction accuracy of the ionomer volume fraction and its gradient distribution by combining physical and data-driven models. Although it involves gradient distribution, it mainly focuses on conventional linear gradients and does not address robust optimization of ionomer content gradients under wide humidity conditions. Therefore, a method oriented towards humidity-varying conditions is urgently needed to improve the rationality and robustness of cathode catalyst layer design, providing sufficient degrees of freedom for gradient control while obtaining an ionomer content gradient scheme applicable to a wide humidity range. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a robust optimization method for ionomer gradient in the cathode catalyst layer of a fuel cell and a fuel cell itself. This method addresses the problems of significant local variations in ionomer content requirements within the cathode catalyst layer under different humidity conditions, insufficient adaptability of single humidity optimization schemes, and limited control capabilities of conventional gradient methods. It employs an asymmetric two-segment power-law curve to characterize the ionomer content gradient distribution within the cathode catalyst layer and possesses the ability to flexibly control the transition balance, overall path, and local enrichment or reduction of functional zones on the proton and oxygen inlet sides. Furthermore, based on the humidity range, it determines the boundary humidity conditions and robust optimization objectives, generates a multi-humidity response dataset reflecting mass transfer bottlenecks under extreme conditions, and performs multi-objective robust optimization to screen for the optimal robust ionomer gradient scheme that balances output performance and reaction uniformity within the humidity range.

[0006] To achieve the above objectives, the technical solution of the present invention is: a robust optimization method for the gradient of ionomers in the cathode catalyst layer of a fuel cell, comprising the following steps: S1. Determine the design objectives, humidity operating range, and design benchmarks for the fuel cell cathode catalyst layer; S2. The gradient distribution of ionomer content in the cathode catalyst layer is characterized by an asymmetric two-segment power-law curve, and the range of values ​​for the control parameters of the asymmetric two-segment power-law curve is determined. S3. Determine the boundary humidity conditions based on the humidity range, and determine the robust optimization objective under the boundary humidity conditions; S4. Based on the control parameters and their value ranges of the asymmetric two-segment power-law curve control parameters, generate a multi-humidity response dataset, construct a surrogate model, and perform multi-objective robust optimization based on the surrogate model to obtain a candidate robust solution set. S5. Select the optimal robust solution from the candidate robust solution set, and verify the optimal robust solution within the humidity range.

[0007] Furthermore, the design objectives include at least one of performance, durability, and cost objectives; the humidity operating range is preferably between 50% and 100% relative humidity; and the design benchmark is the corresponding uniform ionomer distribution scheme obtained by screening according to the design objectives under different humidity conditions.

[0008] Furthermore, the carbon content or carbon support framework parameters in the cathode catalyst layer are kept constant, and the ionomer content distribution is controlled by adjusting the mass ratio of ionomer to carbon support.

[0009] Furthermore, the expression for the asymmetric two-segment power-law curve used to characterize the ionomer content distribution is as follows:

[0010] In the formula: These are the normalized coordinates along the thickness direction of the cathode catalyst layer. Normalized coordinates y norm The local carbon ratio at the location, It is the distance from the transition point to carbon, Δ PEM It is the membrane-side gradient amplitude, Δ MPL It is the gradient amplitude on the microporous layer side. y tran These are the coordinates of the transition point. n PEM It is the membrane-side power exponent. n MPL It is the power exponent of the microporous layer, with a total of six control parameters.

[0011] Furthermore, the asymmetric bisegment power-law curve includes a proton inlet-side functional region near the proton exchange membrane and an oxygen inlet-side functional region near the microporous layer or gas diffusion layer. y tran and Used to control the relative range and transition balance between the two functional areas, ΔPEM and Δ MPL Used to regulate the overall transport support for proton entry on the membrane side and oxygen entry on the microporous layer side. n PEM and n MPL The control parameter is used to control the local enrichment of ionomers on the membrane side and the local reduction of ionomers on the microporous layer side; the range of the control parameter is determined based on at least one of ionomer content constraints, gradient curve morphology constraints, manufacturing constraints, and design target response.

[0012] Furthermore, the boundary humidity conditions include low-humidity boundary conditions and high-humidity boundary conditions within the humidity range, with each boundary condition being 50%. RH Low humidity conditions and 100% RH The operating conditions corresponding to high humidity conditions; the robust optimization objectives include performance objectives under low humidity boundary conditions, performance objectives under high humidity boundary conditions, and durability objectives under boundary humidity conditions; the performance objectives are characterized by peak net output power, and the durability objectives are characterized by the maximum value among the non-uniformity of the reaction within the cathode catalyst layer under multiple boundary humidity conditions.

[0013] Furthermore, the multi-humidity response dataset includes sample combinations of the control parameters, and robust optimization objectives corresponding to each sample combination under multiple boundary humidity conditions; the multi-humidity response dataset is generated through at least one of numerical calculation, experimental testing, and historical operating data.

[0014] Furthermore, the surrogate model is used to characterize the nonlinear mapping relationship between the control parameters and the robust optimization objective, and the surrogate model is a backpropagation neural network model; the multi-objective robust optimization uses a non-dominated sorting genetic algorithm to obtain a candidate robust solution set.

[0015] Furthermore, the optimal robust solution is selected from the candidate robust solution set using the approximation ideal solution sorting method; the verification of the optimal robust solution within the humidity range includes: verifying whether the net output power and response uniformity of the optimal robust solution are improved relative to the design baseline under the boundary humidity condition and at least one intermediate humidity condition.

[0016] Simultaneously, a fuel cell can be provided in which the ionomer content gradient distribution in the cathode catalyst layer is determined by the aforementioned fuel cell cathode catalyst layer ionomer gradient robust optimization method.

[0017] Compared with the prior art, the present invention has the following beneficial effects: This invention uses an asymmetric two-segment power-law curve to characterize the gradient distribution of ionomer content in the cathode catalyst layer. Compared with traditional linear, step-like or symmetric gradient distributions, it can adjust the degree of local enrichment or reduction of ionomers in the functional regions on the proton inlet and oxygen inlet sides, the overall transport path of reactants, and adjust the balance of the functional regions on both sides through transition parameters. This allows for more flexible adaptation to the nonlinear reactant transport requirements within the cathode catalyst layer that vary with relative humidity.

[0018] Furthermore, this invention performs robust optimization for varying humidity conditions, simultaneously considering the dominant mass transfer bottlenecks under both low and high humidity boundary conditions, thus avoiding the insufficient adaptability caused by optimizing only for a single humidity condition. The resulting ionomer gradient scheme maintains good output performance and reaction uniformity under different humidity conditions, demonstrating broad applicability to the variable humidity environment in actual fuel cell operation.

[0019] Furthermore, this invention combines multi-humidity response datasets, surrogate models, and multi-objective robust optimization to transform the complex ionomer gradient distribution optimization problem into a computable and screenable parameter optimization problem. This reduces a large amount of trial-and-error calculations or experimental screening, improves the efficiency of ionomer gradient design, and provides a systematic optimization path for robust design of fuel cell cathode catalyst layers over a wide humidity range.

[0020] Furthermore, this invention can fully leverage the dual regulatory role of ionomers in proton conduction and oxygen transport. Under low humidity conditions, the optimized robust design enhances membrane-side proton conduction and reduces ohmic losses. Under high humidity conditions, it improves oxygen accessibility on the microporous layer or gas diffusion layer side and alleviates concentration loss. Under intermediate humidity conditions, it achieves superior overall performance by balancing and maintaining sufficient proton and oxygen transport capabilities. By incorporating reaction inhomogeneity into the robust optimization and verification process, it can suppress regions with excessively strong or weak local reactions, reduce the risk of reaction homogenization deterioration within the cathode catalyst layer, mitigate local electrochemical load concentration, and improve the operational stability and durability of the fuel cell.

[0021] Furthermore, fuel cells that use the method described in this invention to determine the gradient distribution of ionomers can improve catalyst utilization under low platinum loading conditions, while taking into account net output power, reaction uniformity, and humidity adaptability. This has the potential to reduce the cost of using precious metals and provides a practical and scalable optimization path for low platinum membrane electrode design. Attached Figure Description

[0022] To more intuitively and clearly illustrate the embodiments of this specification, the embodiments will be briefly described below with accompanying drawings. Obviously, the following drawings are only some embodiments of this specification. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0023] Figure 1 This is a schematic diagram of a robust optimization method for the gradient of ionomers in the cathode catalyst layer of a fuel cell according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the linear distribution of the ionomer content gradient along the thickness direction of the cathode catalyst layer, and the coordinates of different transition point positions, the carbon separation ratio of the transition point, the gradient amplitude on the membrane side, the gradient amplitude on the microporous layer side, the power exponent on the membrane side, and the asymmetric two-segment power law curves under the power exponent on the microporous layer side, according to an embodiment of the present invention. Figure 3 This invention describes the process of obtaining the optimal robust solution through multi-objective robust optimization in an embodiment of the invention. It includes schematic diagrams of the carbon separation ratio distribution of the optimal solution under low humidity and high humidity boundary conditions and the optimal robust solution under multiple humidity conditions. The optimal robust solution is verified by comparing the polarization curves, peak net output power and reaction non-uniformity of these solutions under different humidity conditions within the humidity range. Detailed Implementation

[0024] The specific technical solutions in the embodiments of this specification will be analyzed in detail below with reference to the accompanying drawings, providing a more comprehensive description. Obviously, the following description is only a part of the embodiments and not a description of all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] It is hereby stated that, unless otherwise defined, all technical and scientific terms used in the embodiments and accompanying drawings of this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Furthermore, the terms "comprising" and "having," and any variations thereof, used in the embodiments and accompanying drawings of this specification, indicate a non-exclusive inclusion relationship.

[0026] This invention provides a robust optimization method for the gradient of ionomers in the cathode catalyst layer of a fuel cell, such as... Figure 1 As shown, it includes the following steps: Step 1: Determine the design objectives, humidity range, and design benchmarks for the fuel cell cathode catalyst layer; Step 2: Characterize the gradient distribution of ionomer content in the cathode catalyst layer using an asymmetric two-segment power-law curve, and determine the range of values ​​for the control parameters of the asymmetric two-segment power-law curve. Step 3: Determine the boundary humidity conditions based on the humidity range, and determine the robust optimization objective under the boundary humidity conditions; Step 4: Generate a multi-humidity response dataset based on the control parameters and their value ranges, construct a surrogate model, and perform multi-objective robust optimization based on the surrogate model to obtain a candidate robust solution set; Step 5: Select the optimal robust solution from the candidate robust solution set, and verify the optimal robust solution within the humidity range.

[0027] use Figure 1 The robust optimization method for ionomer gradient in the cathode catalyst layer of a fuel cell, as shown, combines a three-dimensional multiphase numerical model of a proton exchange membrane fuel cell (PEMFC) that integrates an agglomerate sub-model, presenting a complete process for optimizing the ionomer content distribution within the cathode catalyst layer (CCL). To isolate the influence of other parameters, the carbon volume fraction and thickness of the CCL are fixed at 0.25 and 10 μm, respectively, and the ionomer content gradient distribution is achieved by adjusting the local ionomer to carbon support mass ratio.

[0028] Step one involves considering the limitations of PEMFCs in terms of performance, durability, and cost. Improving the battery's net output power and the uniformity of the reaction within the CCL are the design goals for performance and durability, respectively; the humidity operating range is 50% relative humidity. RH The humidity conditions between 50% and 100% relative humidity are varied by adjusting the relative humidity of the gas entering the anode and cathode. The preferred design baseline is the uniform ionomer distribution scheme with the highest net power obtained under different humidity conditions. The carbon separation ratios corresponding to the baseline schemes at 50%, 62.5%, 75%, 87.5%, and 100% humidity are 1.20, 1.10, 0.95, 0.80, and 0.70, respectively. Furthermore, to limit costs, all baseline schemes include a platinum loading of 0.1 mg / cm³. 2 The low-platinum cathode catalyst layer.

[0029] In step two, an asymmetric two-segment power-law curve is used to characterize the carbon mass ratio distribution along the thickness direction of the cathode catalyst layer. The expression for the asymmetric two-segment power-law curve is:

[0030] In the formula: These are the normalized coordinates along the thickness direction of the cathode catalyst layer. Normalized coordinates y norm The local carbon ratio at the location, It is the distance from the transition point to carbon, Δ PEM It is the gradient amplitude on the proton exchange membrane (PEM) side, Δ MPL It is the gradient amplitude on the microporous layer (MPL) side.y tran These are the coordinates of the transition point. n PEM It is the membrane-side power exponent. n MPL It is the power exponent of the microporous layer, with a total of six control parameters.

[0031] Figure 2 This is a schematic diagram of the linear distribution of the ionomer content gradient along the thickness direction of the cathode catalyst layer, and the asymmetric bisegment power-law curves under different transition point coordinates, transition point carbon separation ratio, membrane-side gradient amplitude, microporous layer-side gradient amplitude, membrane-side power exponent, and microporous layer-side power exponent, used in embodiments of the present invention. Figure 2 In Figure a, a linear carbon separation ratio distribution is used as an example to illustrate... y tran , Δ PEM and Δ MPL The physical meaning of this is that protons enter the CCL from the PEM side, while oxygen enters from the MPL side. The asymmetric two-segment power-law curve includes a proton inlet-side functional region near the proton exchange membrane and an oxygen inlet-side functional region near the microporous layer or gas diffusion layer. y tran and It is precisely the location coordinates of the transition point between these two functional areas and the carbon separation ratio, such as Figure 2 As shown in Figure b, with y tran 0.5 Using a CCL of 0.95 as a comparative example, adjustments were made. y tran and It can achieve control over the relative range and transition balance between the two functional zones. Given that higher ionomer content is more favorable for proton conduction, while lower ionomer content is more favorable for oxygen transport, Figure c shows that, with Δ... PEM and Δ MPL The CCL values ​​are all 0.5, serving as a comparative example. Δ PEM and Δ MPL It can be used to modulate overall transport support for proton entry on the membrane side and oxygen entry on the microporous layer side. Figure d shows that, with n PEM and n MPL Both CCL values ​​are 2.0, serving as a comparative example. n PEM and n MPL It is used to control the local enrichment of ionomers on the membrane side and the local reduction of ionomers on the microporous layer side, so as to achieve more precise local shape control.

[0032] In this embodiment, to ensure reasonable ionomer content, gradient curve morphology, actual manufacturing, and a clear design target response, y tran The value range is [0.30, 0.70]. The value range is [0.50, 1.25], Δ PEM The value range is [0.30, 0.90], Δ MPL The value range is [0.30, 0.90]. n PEM The value range is [1.00, 2.50]. n MPL The value range is [1.00, 2.50], and an additional constraint is added, namely, the local carbon separation ratio within the entire CCL is not less than 0.05, in order to avoid unrealistic designs.

[0033] Perform step three, based on the selected humidity range, and set 50%... RH Low humidity conditions and 100% RH The high humidity condition was defined as the boundary humidity condition, representing the proton conduction limitation under dry conditions and the oxygen transport limitation under humid conditions, respectively. The robust optimization objective was set for these two boundary conditions, using 50%. RH With 100% RH The peak net output power was used as two performance targets, and in order to limit the abnormal reaction area, 50% was adopted. RH With 100% RH The larger value in the non-uniformity of the reaction within the lower cathode catalyst layer represents the durability target, thus reasonably simplifying this four-dimensional optimization problem into a three-dimensional optimization problem.

[0034] Step four involves generating a multi-humidity response dataset based on the control parameters and their value ranges of the asymmetric two-segment power-law ionomer gradient using the Latin hypercube sampling method. This dataset includes sample combinations of control parameters and the values ​​of each sample combination within a 50% range. RH With 100% RH The three robust optimization objectives derived from simulations were used to generate a dataset containing 120 samples. Due to the need for simulations under two boundary conditions, a total of 240 complete numerical calculations were performed. Subsequently, a backpropagation neural network was trained using this dataset as a surrogate model to represent the nonlinear mapping relationship between the control parameters and the robust optimization objectives. The prediction results showed a high degree of agreement with the simulation results, with the determination coefficients for all three output objectives exceeding 0.99 and the mean square error consistently below 1×10⁻⁶. -4 To increase by 50% RH With 100% RH The peak net power is reduced by 50%. RH With 100% RH The larger value in the reaction inhomogeneity within the lower cathode catalyst layer is used to combine this surrogate model with a second-generation non-dominated sorting genetic algorithm to perform multi-objective robust optimization. The resulting candidate robust solution set is as follows: Figure 3 As shown in Figure a.

[0035] Perform step five. Figure 3 This invention describes the process of obtaining the optimal robust solution through multi-objective robust optimization in an embodiment of the invention. It includes schematic diagrams of the carbon separation ratio distribution of the optimal solution under low-humidity and high-humidity boundary conditions, as well as the optimal robust solution under multiple humidity levels. The optimal robust solution is verified by comparing the polarization curves, peak net output power, and reaction non-uniformity of these solutions under different humidity conditions within the humidity range. The optimal robust solution is selected from the candidate robust solution set using a method that approximates the ideal solution ranking, and the solution with the highest relative closeness is chosen as the optimal design. Figure 3 As shown in Figure a, the closeness of the selected solution is 0.6485, which is 50%. RH The lower peak net power is 0.8663W, 100%. RH The lower peak net power is 0.7866 W, while the relatively high reaction uniformity under boundary conditions is limited to 0.5334 W. The corresponding control parameters for the asymmetric two-segment power-law curve are... y tran =0.4921, =0.8508, Δ PEM =0.7618, Δ MPL =0.4951, n PEM =1.4212, and n MPL =1.9296. For comparison, optimizations were also performed at specific humidity levels of 50% and 100% to obtain the optimal single-humidity scheme for maximizing peak net power and minimizing reaction uniformity. The carbon separation ratio distributions of these three schemes are shown in... Figure 3 In Figure b, 50% RH The optimal solution exhibits the widest PEM-side functional region, the highest transition value, and the most significant PEM-side ionomer enrichment, meeting the need to alleviate proton transport limitations under dry conditions. In contrast, 100% RH The optimal solution shows the minimum y tran and This enhances oxygen intake on the MPL side, alleviating the oxygen transport limitation dominated by fully humidified conditions. The optimal robust scheme for multi-humidity operation lies between these two extreme cases, indicating that robust operation requires a balance between proton conduction on the PEM side and oxygen transport on the MPL side.

[0036] The optimal robust solution was validated within a specific humidity range; in this embodiment, a humidity level of 50% was selected. RH With 100% RH The boundary conditions and the 75% of intermediate conditions not involved in optimization. RH Under operating conditions, verify whether this optimal robust scheme improves net output power and response uniformity relative to the design baseline. Figure 3 Figures c, d, and e in the middle section compare the polarization curves of fuel cells using three optimized schemes with the baseline scheme under various humidity conditions. 50% RH Optimal solution and 100% RH The optimal solutions performed best under their respective target humidity conditions, achieving net peak power improvements of 4.03% and 4.32%, respectively, compared to the corresponding baseline solutions. However, these are specific to certain conditions. RH The optimized scheme did not perform ideally under other humidity conditions. In contrast, while the multi-humidity optimal robust scheme did not have an absolute advantage under extreme humidity conditions, its peak net power was significantly better than the corresponding best benchmark scheme within the investigated humidity range, at 50%, 62.5%, 75%, 87.5%, and 100%. RH The percentage increases were 3.30%, 3.99%, 4.39%, 4.11%, and 3.55%, respectively. This indicates that the ionomer gradient determined by the robust optimization method can maintain stable performance improvement over a wide humidity range, rather than being effective only under a single operating condition.

[0037] Figure 3 Figures f, g, and h further fairly evaluated the net power and reaction uniformity of these schemes under the same current density, further confirming that the multi-humidity optimal robust scheme maintains a significantly superior performance compared to the corresponding benchmark scheme across the entire humidity range. Under extreme dry and humid conditions, it maintains sufficient proton conduction on the PEM side and oxygen transport on the MPL side, respectively, thereby alleviating the dominant mass transfer bottleneck under boundary humidity conditions and achieving close to 100% and 50% respectively. RH The net work done by the optimal solution. And at 75%... RH Under these conditions, the optimal robust scheme achieves the maximum net power gain of 4.71% by simultaneously increasing both overall proton conductivity and platinum surface oxygen concentration, thereby limiting ohmic and concentration losses, although this condition was not directly incorporated into the optimization objective. This transport balance also prevents excessively high local reaction rates, effectively limiting the increase in reaction non-uniformity over a wide humidity range, and within 50%... RH It even exhibits the lowest non-uniformity. These results indicate that the optimal robust scheme is not a simple average of the optimal schemes under low and high humidity conditions, but rather a balanced scheme that achieves broad humidity applicability by maintaining sufficient proton and oxygen transport.

[0038] The preferred embodiments of the present invention have been described above by way of illustration only and are not intended to limit the present invention. Those skilled in the art can make various modifications and variations to the foregoing embodiments. Therefore, any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A robust optimization method for the gradient of ionomers in the cathode catalyst layer of a fuel cell, characterized in that, Includes the following steps: S1. Determine the design objectives, humidity operating range, and design benchmarks for the fuel cell cathode catalyst layer; S2. The gradient distribution of ionomer content in the cathode catalyst layer is characterized by an asymmetric two-segment power-law curve, and the range of values ​​for the control parameters of the asymmetric two-segment power-law curve is determined. S3. Determine the boundary humidity conditions based on the humidity range, and determine the robust optimization objective under the boundary humidity conditions; S4. Based on the control parameters and their value ranges of the asymmetric two-segment power-law curve control parameters, generate a multi-humidity response dataset, construct a surrogate model, and perform multi-objective robust optimization based on the surrogate model to obtain a candidate robust solution set. S5. Select the optimal robust solution from the candidate robust solution set, and verify the optimal robust solution within the humidity range.

2. The robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to claim 1, characterized in that, The design objectives include at least one of performance, durability and cost objectives; the humidity range is preferably between 50% and 100% relative humidity; the design benchmark is the corresponding uniform ionomer distribution scheme obtained by screening according to the design objectives under different humidity conditions.

3. The robust optimization method for the ionomer gradient of the fuel cell cathode catalyst layer according to claim 1, characterized in that, Keep the carbon content or carbon support framework parameters in the cathode catalyst layer constant, and control the ionomer content distribution by adjusting the mass ratio of ionomer to carbon support.

4. The robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to claim 1, characterized in that, The expression for the asymmetric two-segment power-law curve used to characterize the ionomer content distribution is as follows: In the formula: These are the normalized coordinates along the thickness direction of the cathode catalyst layer. Normalized coordinates y norm The local carbon ratio at the location, It is the distance from the transition point to carbon, Δ PEM It is the membrane-side gradient amplitude, Δ MPL It is the gradient amplitude on the microporous layer side. y tran These are the coordinates of the transition point. n PEM It is the membrane-side power exponent. n MPL It is the power exponent of the microporous layer, with a total of six control parameters.

5. The robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to claim 4, characterized in that, The asymmetric bisegment power-law curve includes a proton inlet-side functional region near the proton exchange membrane and an oxygen inlet-side functional region near the microporous layer or gas diffusion layer. y tran and Used to control the relative range and transition balance between the two functional areas, Δ PEM and Δ MPL Used to regulate the overall transport support for proton entry on the membrane side and oxygen entry on the microporous layer side. n PEM and n MPL The control parameter is used to control the local enrichment of ionomers on the membrane side and the local reduction of ionomers on the microporous layer side; the range of the control parameter is determined based on at least one of ionomer content constraints, gradient curve morphology constraints, manufacturing constraints, and design target response.

6. The robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to claim 1, characterized in that, The boundary humidity conditions include low-humidity boundary conditions and high-humidity boundary conditions within the humidity range, with each condition representing 50%. RH Low humidity conditions and 100% RH Operating conditions corresponding to high humidity; The robust optimization objectives include performance objectives under low humidity boundary conditions, performance objectives under high humidity boundary conditions, and durability objectives under boundary humidity conditions. The performance objectives are characterized by peak net output power, and the durability objectives are characterized by the maximum value among the non-uniformity of the reaction within the cathode catalyst layer under multiple boundary humidity conditions.

7. The robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to claim 1, characterized in that, The multi-humidity response dataset includes sample combinations of the control parameters and robust optimization objectives corresponding to each sample combination under multiple boundary humidity conditions; the multi-humidity response dataset is generated through at least one of numerical calculation, experimental testing, and historical operating data.

8. The robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to claim 1, characterized in that, The surrogate model is used to characterize the nonlinear mapping relationship between the control parameters and the robust optimization objective. The surrogate model is a backpropagation neural network model. The multi-objective robust optimization uses a non-dominated sorting genetic algorithm to obtain a candidate robust solution set.

9. The robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to claim 1, characterized in that, The optimal robust solution is selected from the candidate robust solution set using the approximation ideal solution sorting method; the optimal robust solution is verified within the humidity range, including: verifying whether the net output power and response uniformity of the optimal robust solution are improved relative to the design baseline under the boundary humidity condition and at least one intermediate humidity condition.

10. A fuel cell, characterized in that, The cathode catalyst layer includes a cathode catalyst layer, wherein the gradient distribution of ionomer content in the cathode catalyst layer is determined by the robust optimization method for ionomer gradient of fuel cell cathode catalyst layer according to any one of claims 1-9.

Images