Methods, apparatus, equipment and storage media for determining parameters of gas diffusion layers

Abstract

The present invention discloses a method, device, equipment and storage medium for determining parameters of a gas diffusion layer, which relates to the field of fuel cell technology and solves the problem that the determination of parameters of a gas diffusion layer requires a large amount of economic cost and time cost. The specific scheme includes: obtaining a simulation model of the gas diffusion layer, the simulation model includes simulation parameters; obtaining parameter values ​​corresponding to the simulation parameters; performing simulation calculations on the simulation model and the parameter values ​​corresponding to the simulation parameters to obtain a first simulation result; if the simulation result is used to indicate a simulation failure, the prompt information in the first simulation result is input into a trained parameter revision model to obtain a revised parameter value; performing simulation calculations on the simulation model and the revised parameter value to obtain a second simulation result; if the second simulation result is used to indicate a simulation success, the revised parameter value is determined as the target parameter value.

Description

Technical Field

[0001] The present invention relates to the field of fuel cell technology, and in particular to a method, device, equipment and storage medium for determining parameters of a gas diffusion layer. Background Art

[0002] The gas diffusion layer plays an important role in supporting the catalyst layer, collecting current, conducting gas and discharging the reaction product water in the fuel cell. As an important component structure in the proton exchange membrane fuel cell, the gas diffusion layer accounts for 20-25% of the cost of the proton exchange membrane fuel cell. Its performance directly affects whether the fuel cell can work normally. The determination of the parameters of the gas diffusion layer is a key step in preparing a gas diffusion layer with excellent performance.

[0003] In the prior art, in order to determine the parameters of a gas diffusion layer with excellent performance, it is usually necessary to prepare the gas diffusion layer through actual experiments and obtain the parameters of the gas diffusion layer with excellent performance through testing with specific testing instruments. However, this method is costly and the experimental preparation and testing time is relatively long. Summary of the invention

[0004] The present application provides a method, device, equipment and storage medium for determining parameters of a gas diffusion layer, which solves the problem that determining the parameters of the gas diffusion layer requires a large amount of economic cost and time cost.

[0005] In order to achieve the above objectives, this application adopts the following technical solutions:

[0006] According to a first aspect of an embodiment of the present application, a method for determining parameters of a gas diffusion layer is provided, the method comprising: obtaining a simulation model of the gas diffusion layer, the simulation model comprising simulation parameters; obtaining parameter values ​​corresponding to the simulation parameters; performing simulation calculations on the simulation model and the parameter values ​​corresponding to the simulation parameters to obtain a first simulation result; if the simulation result is used to indicate a simulation failure, inputting prompt information in the first simulation result into a trained parameter revision model to obtain a revised parameter value; performing simulation calculations on the simulation model and the revised parameter value to obtain a second simulation result; if the second simulation result is used to indicate a simulation success, determining the revised parameter value as the target parameter value.

[0007] In one embodiment, obtaining a parameter value corresponding to a simulation parameter includes:

[0008] The simulation model is input into the trained first model to obtain the value range of the parameter value corresponding to each simulation parameter, and the parameter value of each simulation parameter is determined from the value range to obtain the parameter value corresponding to the simulation parameter.

[0009] In one embodiment, the second simulation result includes: a compression ratio of the gas diffusion layer and a displacement corresponding to the compression ratio;

[0010] If the second simulation result indicates that the simulation is successful, determining the revised parameter value as the target parameter value includes:

[0011] If the second simulation result is used to indicate that the simulation is successful, when it is determined that the displacement corresponding to the compression ratio in the second simulation result is within a preset range, the revised parameter value is determined as the target parameter value.

[0012] In one embodiment, the method further comprises:

[0013] Obtain the compression ratio of the simulation parameters under different target parameter values, and the displacement corresponding to each compression ratio;

[0014] According to the displacement changes under different compression ratios, sensitive parameters in the simulation parameters are determined. The sensitive parameters are used to indicate the parameters that have a greater impact on the displacement changes.

[0015] In one embodiment, the simulation parameters include a simulation step size and a calculation domain, and the method further includes:

[0016] Determine the calculation time of the simulation model according to the simulation step size and the calculation domain;

[0017] Perform simulation calculations on parameter values ​​corresponding to the simulation model and simulation parameters, including:

[0018] If the calculation time is less than a preset threshold, simulation calculation is performed on the parameter values ​​corresponding to the simulation model and the simulation parameters.

[0019] In one embodiment, before obtaining the simulation model of the gas diffusion layer, the method further includes:

[0020] Obtain different simulation models of multiple gas diffusion layers and prompt information of each simulation model;

[0021] The parameter revision model is trained according to different simulation models and prompt information of each simulation model to obtain a trained parameter revision model.

[0022] In one embodiment, before inputting the simulation model into the trained first model, the method further comprises:

[0023] Acquire different simulation models of multiple gas diffusion layers and target parameter values ​​corresponding to each simulation model;

[0024] The parameter revision model is trained according to different simulation models and target parameter values ​​corresponding to each simulation model to obtain a trained first model.

[0025] According to a second aspect of the embodiment of the present application, a device for determining parameters of a gas diffusion layer is further provided, the device comprising:

[0026] An acquisition module is used to acquire a simulation model of a gas diffusion layer, where the simulation model includes simulation parameters and parameter values ​​corresponding to the simulation parameters;

[0027] A first simulation module, used to perform simulation calculation on parameter values ​​corresponding to the simulation model and the simulation parameters to obtain a first simulation result;

[0028] A processing module, configured to input the prompt information in the first simulation result into the trained parameter revision model to obtain a revised parameter value if the simulation result indicates that the simulation fails;

[0029] A second simulation module is used to perform simulation calculation on the simulation model and the revised parameter value to obtain a second simulation result;

[0030] The determination module is configured to determine the revised parameter value as the target parameter value if the second simulation result indicates that the simulation is successful.

[0031] According to a third aspect of an embodiment of the present application, a computer device is provided, which includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, any one of the gas diffusion layer parameter determination methods provided in the first aspect of the embodiment of the present application is implemented.

[0032] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any gas diffusion layer parameter determination method provided in the first aspect of the embodiment of the present application can be implemented.

[0033] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least:

[0034] The method for determining the parameters of the gas diffusion layer provided in the embodiment of the present application obtains the simulation model of the gas diffusion layer and obtains the parameter values ​​corresponding to the simulation parameters, performs simulation calculation on the simulation model and the parameter values ​​corresponding to the simulation parameters, and when the simulation calculation fails, inputs the prompt information into the trained parameter revision model to obtain the revised parameter value, performs simulation calculation on the simulation model and the revised parameter value, and when the simulation calculation succeeds, determines the revised parameter value as the target parameter value. The method for determining the parameters of the gas diffusion layer provided in the embodiment of the present application determines the parameters of the gas diffusion layer with excellent performance by the method of simulation calculation. Compared with the method of determining the parameters of the gas diffusion layer by experimental preparation and testing, this method can save the economic cost and time cost of parameter determination. At the same time, the method for determining the parameters of the gas diffusion layer provided in the embodiment of the present application can quickly revise the parameters by the parameter revision model when the simulation fails, which can further improve the efficiency of determining the parameters of the gas diffusion layer. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram of the internal structure of a computer device provided by an embodiment of the present invention;

[0036] Figure 2 A flow chart of a method for determining gas diffusion layer parameters provided by an embodiment of the present invention;

[0037] Figure 3 A structural diagram of a gas diffusion layer parameter determination device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, "plurality" means two or more.

[0040] Additionally, the use of “based on” or “according to” is meant to be open and inclusive, as a process, step, calculation, or other action “based on” or “according to” one or more conditions or values ​​may, in practice, be based on additional conditions or beyond values.

[0041] Fuel cell vehicles are a type of modern electric vehicles. Their body, powertrain, and control system are basically the same as those of pure electric vehicles. The main difference is the power battery. Pure electric vehicles use secondary batteries such as lead-acid batteries, nickel-metal hydride batteries, and lithium-ion batteries as power sources, while fuel cell vehicles use fuel cells as power sources, and all power loads are borne by fuel cells. The research, development, and utilization of fuel cell vehicles have always been highly valued. Although the life, cost, and performance of automotive fuel cells need to be further optimized, preparations for the industrialization of fuel cell vehicles have begun, and some have even announced that they have entered the mass production stage. Fuel cells will become the fourth generation of power generation after thermal power, hydropower, and nuclear power. According to the type of fuel cell, fuel cells are mainly divided into alkaline fuel cells, phosphoric acid fuel cells, molten carbonate fuel cells, solid oxide fuel cells, direct methanol fuel cells, and proton exchange membrane fuel cells.

[0042] Proton exchange membrane fuel cells are the main type of fuel cells at present. The fuel cell stack is the power source of the proton exchange membrane fuel cell and is one of the core components of the proton exchange membrane fuel cell. The fuel cell stack mainly includes: collector plate, bipolar plate, gas diffusion layer, microporous layer, catalyst layer and proton exchange membrane, among which the proton exchange membrane, catalyst layer, microporous layer and gas diffusion layer are collectively referred to as membrane electrode assembly.

[0043] Proton exchange membrane batteries have become one of the mainstream technologies for fuel cell application and promotion due to their higher energy density, lower operating temperature, and short start-up time. The gas diffusion layer is one of the important components of proton exchange membrane fuel cells, and its stability has an important influence on the performance and durability of proton exchange membrane fuel cells. In the design of fuel cell stacks, the selection of gas diffusion layers has a great influence on the performance of fuel cells. Usually, a comprehensive trade-off and consideration is made in terms of thickness, specific gravity, compression rebound, thickness, porosity, electrical conductivity, thermal conductivity, and gas diffusion characteristics. The gas diffusion layer has a complex, uneven, and anisotropic microstructure. The changes in the component microstructure caused by the assembly clamping force affect the thickness, porosity, conductivity, and permeability of the gas diffusion layer, and the transport characteristics of the gas diffusion layer change significantly. Improving the durability of the gas diffusion layer is of great significance to the extension of the life of the proton exchange membrane fuel cell. Therefore, the study of the gas diffusion layer is of great significance.

[0044] The gas diffusion layer is usually made of carbon paper or carbon cloth. In order to make the fuel cell as thin as possible, carbon paper is a good preparation material. Carbon paper is made by pyrolyzing non-woven carbon fiber sheets and has good electrical conductivity. However, due to its thin sheet structure, it is often easy to break and fragile, so the handling of carbon paper requires extra caution. When the gas diffusion layer is compressed by the bipolar plate, due to the fragility of carbon paper, if the compression of the gas diffusion layer is too large, it will directly cause damage to the carbon paper and cause losses; if the compression is too small, it will cause gas leakage, and the electrical conductivity will not meet the requirements, affecting the quality of the gas diffusion layer. Therefore, in-depth research on the stress-deformation of the compressed microstructure of the gas diffusion layer, and exploring the influence of changes in the structural parameters of the gas diffusion layer before and after compression on the gas-water-heat-electricity transmission characteristics have become the key to preparing high-quality carbon paper.

[0045] In the prior art, in order to determine the parameters of a gas diffusion layer with excellent performance, it is usually necessary to prepare the gas diffusion layer through actual experiments and obtain the parameters of the gas diffusion layer with excellent performance through testing with specific testing instruments. However, this method is costly and the experimental preparation and testing time is relatively long.

[0046] In order to solve the above problems, the embodiment of the present application provides a method for determining the parameters of a gas diffusion layer, by obtaining a simulation model of the gas diffusion layer and obtaining parameter values ​​corresponding to the simulation parameters, performing simulation calculations on the simulation model and the parameter values ​​corresponding to the simulation parameters, and when the simulation calculation fails, inputting prompt information into a trained parameter revision model to obtain revised parameter values, performing simulation calculations on the simulation model and the revised parameter values, and when the simulation calculation succeeds, determining the revised parameter values ​​as target parameter values. The method for determining the parameters of a gas diffusion layer provided in the embodiment of the present application determines the parameters of a gas diffusion layer with excellent performance by means of simulation calculations, and compared with determining the parameters of a gas diffusion layer by means of experimental preparation and testing, this method can save the economic cost and time cost of parameter determination.

[0047] Furthermore, the method for determining the parameters of the gas diffusion layer provided in the embodiment of the present application can quickly revise the parameters through a parameter revision model when the simulation fails, which can further improve the efficiency of determining the parameters of the gas diffusion layer.

[0048] The execution subject of the gas diffusion layer parameter determination method provided in the embodiment of the present application can be a computer device, a terminal device, or a server, wherein the terminal device can be various personal computers, laptops, smart phones, tablet computers, portable wearable devices, etc., and the present application does not make specific limitations.

[0049] Figure 1 The internal structure diagram of a computer device provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the computer device includes a processor and a memory connected via a system bus. The processor is used to provide computing and control capabilities. The memory may include a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The computer program can be executed by the processor to implement the steps of a method for determining gas diffusion layer parameters provided in each of the above embodiments. The internal memory provides a cache operating environment for the operating system and the computer program in the non-volatile storage medium.

[0050] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0051] Based on the above execution subject, an embodiment of the present application provides a method for determining parameters of a gas diffusion layer.

[0052] like Figure 2As shown, the method comprises the following steps:

[0053] Step 201: Obtain a simulation model of a gas diffusion layer.

[0054] The simulation model includes simulation parameters, and the simulation model can be determined according to the actual structure of the gas diffusion layer.

[0055] Specifically, the simulation model can be obtained by using three-dimensional drawing software and drawing the simulation model of the gas diffusion layer according to the actual structure of the gas diffusion layer, or the simulation model of the gas diffusion layer can be directly imported from a local computer.

[0056] Step 202: Obtain parameter values ​​corresponding to simulation parameters;

[0057] There are multiple simulation parameters, for example, the simulation parameters may be: calculation step, calculation domain, Young's modulus, Poisson's ratio, density, compression speed, mass scaling factor, etc. In the actual simulation calculation process, it is necessary to assign values ​​to each simulation parameter before starting the simulation calculation.

[0058] Specifically, the basis for selecting the above simulation parameters may be:

[0059] (1) Time step: The time step is very important and plays a key role in the calculation results. If the step is not chosen well, not only will the curve be not smooth, but the simulation may not converge. (2) Computational domain: The choice of the computational domain should cover the entire process of the event. If the computational domain is too small, accurate simulation results cannot be obtained, and the simulation may not converge; if the computational domain is too large, the calculation time will be too long, resources will be wasted, and the cost will increase. (3) Young's modulus: If the Young's modulus is too large, the gas diffusion layer will be subjected to excessive force, making it difficult to compress and the simulation will be interrupted. (4) Poisson's ratio: It affects the degree of fiber deformation. Fibers with a small Poisson's ratio have a low deformation. (5) Density: If the density is too high, the system damping will increase, the simulation calculation time will increase, and the simulation may be interrupted and the calculation will stop. (6) Compression speed: For the gas diffusion layer, the compression process is the key. Different compression speeds will generate different inertial forces, which will affect the simulation results. The compression speed is also a key data for the actual preparation of carbon paper. Therefore, it is particularly important to obtain an appropriate compression speed during simulation. (7) Mass scaling factor: The hazards of an inappropriate mass scaling factor are: excessive inertial force due to mass scaling, resulting in distorted results; excessive unit kinetic energy due to mass scaling, which may cause the unit to deform too quickly, or even exceed the wave speed and terminate the calculation. Using an appropriate mass scaling factor can greatly improve the calculation efficiency without losing too much model accuracy, saving calculation costs.

[0060] Step 203: perform simulation calculation on the simulation model and the parameter values ​​corresponding to the simulation parameters to obtain a first simulation result.

[0061] The first simulation result may indicate a simulation success or a simulation failure. If the simulation is successful, the simulation result includes displacement data of the gas diffusion layer at different compression ratios. If the simulation fails, prompt information of the simulation failure may be obtained from the simulation result, such as the simulation failure caused by unreasonable settings of certain specific parameters, or the simulation failure caused by insufficient computing power of the device to enable the simulation terminal.

[0062] Step 204: If the first simulation result indicates a simulation failure, the prompt information in the first simulation result is input into the trained parameter revision model to obtain a revised parameter value.

[0063] It should be noted that in the actual simulation process, the simulation often fails due to unreasonable parameter settings, or due to errors in simulation calculations caused by the model being too large, etc. If the simulation result obtained indicates that the simulation has failed, the prompt information in the simulation result is input into the trained parameter revision model to obtain the revised parameter value.

[0064] The prompt information may include the type of simulation failure or prompt information of an unreasonable parameter setting. The prompt information may be input into a trained parameter revision model to assist in parameter revision.

[0065] Step 205: perform simulation calculation on the simulation model and the revised parameter values ​​to obtain a second simulation result.

[0066] Step 206: If the second simulation result indicates that the simulation is successful, the revised parameter value is determined as the target parameter value.

[0067] In the actual execution process, after obtaining the revised parameter value, the simulation model is re-simulated and calculated, and a second simulation result is obtained. If the second simulation result indicates that the simulation is successful, the revised parameter value is determined as the target parameter value. If the second simulation result indicates that the simulation fails, the prompt information in the simulation result is re-entered into the parameter revision model to revise the parameters again until the simulation is successful.

[0068] Optionally, obtaining parameter values ​​corresponding to simulation parameters includes: inputting the simulation model into a trained first model, obtaining a value range of parameter values ​​corresponding to each simulation parameter, and determining the parameter value of each simulation parameter from the value range to obtain the parameter value corresponding to the simulation parameter.

[0069] The first model is a model based on an expert database, and the range of parameter values ​​of the simulation parameters of the simulation model can be obtained by inputting the simulation model into the model.

[0070] When the parameter values ​​corresponding to the simulation parameters are determined for the first time, the simulation model can be input into the trained first model, and the parameter values ​​of the simulation parameters corresponding to the simulation model can be obtained through the first model.

[0071] In one embodiment, the second simulation result includes: a compression ratio of the gas diffusion layer and a displacement corresponding to the compression ratio; if the second simulation result is used to indicate a successful simulation, the revised parameter value is determined as the target parameter value, including: if the second simulation result is used to indicate a successful simulation, when it is determined that the displacement corresponding to the compression ratio in the second simulation result is within a preset range, the revised parameter value is determined as the target parameter value.

[0072] It should be noted that when the simulation results indicate that the simulation is successful, it is also necessary to obtain the compression ratio of the gas diffusion layer and the displacement corresponding to the compression ratio in the simulation results, and determine whether the displacement under different compression ratios is within the preset value range. If so, the revised parameter value is determined as the target parameter value. If not, it is necessary to re-obtain the parameter values ​​of the simulation parameters of the simulation model and re-simulate.

[0073] Specifically, the displacement range of the gas diffusion layer under different compression ratios can be: the displacement range of the gas diffusion layer at an 8% compression ratio is: -10 to +40 microns, the displacement range at a 17% compression ratio is: -15 to +80 microns, and the displacement range at a 30% compression ratio is: -20 to +120 microns.

[0074] In one embodiment, the method further includes: obtaining the compression ratio of the simulation parameters under different target parameter values, and the displacement corresponding to each compression ratio; and determining the sensitive parameters in the simulation parameters according to the displacement changes under different compression ratios.

[0075] Among them, sensitive parameters are used to indicate parameters that have a greater impact on displacement changes.

[0076] It should be noted that the determination of sensitive parameters can play a guiding role in the actual preparation process of the gas diffusion layer. For example, if a certain parameter is determined to be a sensitive parameter based on the simulation results, then in the actual preparation process, the preparation accuracy of the parameter should be guaranteed, or the impact of the change of the parameter on the performance of the prepared gas diffusion layer should be considered.

[0077] For example, the distribution of the number of nodes with different displacement values ​​in the Z direction can be obtained by using the same simulation model under 5 examples and with a compression ratio of 20%. According to the displacement changes caused by different examples, it can be determined that the friction coefficient in the simulation parameters is a sensitive parameter, while the density, Young's modulus and Poisson's ratio are non-sensitive parameters.

[0078] The specific parameter values ​​of the specific simulation model are shown in Table 1.

[0079] Table 1 Different values ​​of gas diffusion layer simulation parameters

[0080] Calculation example Friction coefficient Young's modulus / [Pa] Poisson's ratio Density / [kg·m3] Bipolar Plate 1.97×1011 0.3 7800 Microplate 2.21×108 0.26 2059 Calculation example 1 0.1 6.1×106 0.1 440 Calculation example 2 0.7 6.1×106 0.1 440 Calculation example 3 0.1 6.1×108 0.1 440 Calculation example 4 0.1 6.1×106 0.256 440 Calculation example 5 0.1 6.1×106 0.1 1000

[0081] In one embodiment, the simulation parameters include a simulation step and a calculation domain, and the method further includes: determining the calculation time of the simulation model based on the simulation step and the calculation domain; performing simulation calculations on parameter values ​​corresponding to the simulation model and the simulation parameters, including: if the calculation time is less than a preset threshold, performing simulation calculations on the parameter values ​​corresponding to the simulation model and the simulation parameters.

[0082] It should be noted that since the simulation model of the gas diffusion layer is relatively complex, the simulation time is relatively long. At the same time, the simulation step size and the size of the computational domain are also key factors affecting the simulation calculation time. Therefore, the simulation calculation time of the simulation model can be determined according to the simulation step size and the computational domain. If the calculated simulation time is less than the preset threshold, it means that the simulation calculation can be performed. If the simulation time is greater than the preset threshold, it means that the simulation model, simulation step size and computational domain are unreasonable and need to be readjusted.

[0083] Optionally, before obtaining the simulation model of the gas diffusion layer, the method also includes: obtaining different simulation models of multiple gas diffusion layers and prompt information of each simulation model; training the parameter revision model according to the different simulation models and the prompt information of each simulation model to obtain a trained parameter revision model.

[0084] Specifically, the parameter revision model can be a model based on a deep Q network. During the training process of the deep Q network model, the parameters of the current value network (time step, degree of deformation, etc.) are obtained by interacting with the simulation software environment to obtain initial values. After each N rounds of iteration, the parameters of the current value network are copied to the target value network (time step, degree of deformation, etc.). The network parameters are updated by minimizing the mean square error between the current Q value and the target Q value. At the same time, the experience replay mechanism is used to store each transfer sample obtained by interacting with the environment into a replay memory unit. During training, a small batch of samples are randomly extracted from the replay memory unit each time, and the stochastic gradient descent algorithm is used to update the current value and target value network parameters.

[0085] Optionally, before inputting the simulation model into the trained first model, the method also includes: obtaining different simulation models of multiple gas diffusion layers and target parameter values ​​corresponding to each simulation model; training the parameter revision model according to the different simulation models and the target parameter values ​​corresponding to each simulation model to obtain the trained first model.

[0086] Specifically, the first model can be a model based on an expert database, which introduces experts into the learning loop of the agent through the TAMER (training an agent manually via evaluative reinforcement) learning framework, provides reinforcement signals to the agent, and then models the signal through supervised learning to form an expert database. The specific method is: the problems and difficulties to be solved in the gas diffusion layer compression simulation process correspond to the environmental factors in TAMER, and the expert intervenes in the agent learning loop, and the state in each environment is reinforced by the expert guidance and allowed to learn by the agent, and then reinforced prediction is performed through the reinforcement model, and finally the action is selected to be fed back to the environmental factors, and it is continuously iterated, so that the agent can learn more quickly and form an expert database model.

[0087] The present application provides a method for determining parameters of a gas diffusion layer, which obtains a simulation model of the gas diffusion layer and obtains parameter values ​​corresponding to the simulation parameters, performs simulation calculations on the simulation model and the parameter values ​​corresponding to the simulation parameters, and when the simulation calculation fails, inputs prompt information into a trained parameter revision model to obtain revised parameter values, performs simulation calculations on the simulation model and the revised parameter values, and when the simulation calculation succeeds, determines the revised parameter values ​​as target parameter values. The method for determining parameters of a gas diffusion layer provided in the present application determines the parameters of a gas diffusion layer with excellent performance by means of simulation calculations, and compared with determining the parameters of a gas diffusion layer by means of experimental preparation and testing, this method can save the economic cost and time cost of parameter determination.

[0088] At the same time, the deep Q network reinforcement learning training algorithm is applied to the gas diffusion layer simulation calculation, which provides a new solution and idea for studying the fuel cell gas diffusion layer in the assembly process using simulation experiments. And by establishing a gas diffusion layer simulation expert database, it lays the foundation for industry research and development, which can greatly shorten the learning cycle of artificial intelligence in the field of gas diffusion layers. Furthermore, according to the simulation parameters revised by the reinforcement learning training algorithm, the sensitivity analysis of the gas diffusion layer parameters can be carried out, and the influence of each parameter of the gas diffusion layer can be analyzed more deeply, which can speed up the development progress of the gas diffusion layer.

[0089] In addition, this method can autonomously revise the simulation parameters of the gas diffusion layer through the deep Q network reinforcement learning training algorithm, and can obtain the optimization parameters such as the time step, calculation domain, Young's modulus, Poisson's ratio, density, compression speed, and mass scaling factor when assembling the gas diffusion layer simulation, which solves the problem of simulation difficulties and can further replace experiments, thereby solving the problem of huge financial resources consumed by the laboratory observation method of the gas diffusion layer. At the same time, it avoids unnecessary errors caused by observation with the human eye, and uses simulation to obtain more accurate digital results.

[0090] like Figure 3 As shown, the embodiment of the present application also provides a device for determining parameters of a gas diffusion layer, the device comprising:

[0091] An acquisition module 11 is used to acquire a simulation model of a gas diffusion layer, where the simulation model includes simulation parameters and parameter values ​​corresponding to the simulation parameters;

[0092] A first simulation module 12 is used to perform simulation calculation on the simulation model and the parameter values ​​corresponding to the simulation parameters to obtain a first simulation result;

[0093] The processing module 13 is used for inputting the prompt information in the first simulation result into the trained parameter revision model to obtain the revised parameter value if the simulation result indicates that the simulation fails;

[0094] The second simulation module 14 is used to perform simulation calculation on the simulation model and the revised parameter value to obtain a second simulation result;

[0095] The determination module 15 is configured to determine the revised parameter value as the target parameter value if the second simulation result indicates that the simulation is successful.

[0096] In one embodiment, the acquisition module 11 is specifically configured to:

[0097] The simulation model is input into the trained first model to obtain the value range of the parameter value corresponding to each simulation parameter, and the parameter value of each simulation parameter is determined from the value range to obtain the parameter value corresponding to the simulation parameter.

[0098] In one embodiment, the second simulation result includes: the compression ratio of the gas diffusion layer and the displacement corresponding to the compression ratio; the determination module 15 is specifically used for:

[0099] If the second simulation result is used to indicate that the simulation is successful, when it is determined that the displacement corresponding to the compression ratio in the second simulation result is within a preset range, the revised parameter value is determined as the target parameter value.

[0100] In one embodiment, the acquisition module 11 is further used for:

[0101] Obtain the compression ratio of the simulation parameters under different target parameter values, and the displacement corresponding to each compression ratio;

[0102] The determination module is also used to determine sensitive parameters in the simulation parameters according to the displacement changes under different compression ratios. The sensitive parameters are used to indicate parameters that have a greater impact on the displacement changes.

[0103] In one embodiment, the simulation parameters include a simulation step size and a calculation domain, and the determination module 15 is further used to:

[0104] Determine the calculation time of the simulation model according to the simulation step size and the calculation domain;

[0105] The first simulation module 12 is also used for:

[0106] If the calculation time is less than a preset threshold, simulation calculation is performed on the parameter values ​​corresponding to the simulation model and the simulation parameters.

[0107] In one embodiment, the apparatus further comprises a training module 16, wherein the training module 16 is configured to:

[0108] Obtain different simulation models of multiple gas diffusion layers and prompt information of each simulation model;

[0109] The parameter revision model is trained according to different simulation models and prompt information of each simulation model to obtain a trained parameter revision model.

[0110] In one embodiment, the training module 16 is further used to:

[0111] Acquire different simulation models of multiple gas diffusion layers and target parameter values ​​corresponding to each simulation model;

[0112] The parameter revision model is trained according to different simulation models and target parameter values ​​corresponding to each simulation model to obtain a trained first model.

[0113] The embodiment of the present application provides a parameter determination device for a gas diffusion layer, which obtains a simulation model of the gas diffusion layer and obtains parameter values ​​corresponding to the simulation parameters, performs simulation calculations on the simulation model and the parameter values ​​corresponding to the simulation parameters, and when the simulation calculation fails, inputs prompt information into a trained parameter revision model to obtain revised parameter values, performs simulation calculations on the simulation model and the revised parameter values, and when the simulation calculation succeeds, determines the revised parameter values ​​as target parameter values. The parameter determination device for a gas diffusion layer provided in the embodiment of the present application determines the parameters of a gas diffusion layer with excellent performance by a simulation calculation method, and compared with determining the parameters of a gas diffusion layer by an experimental preparation and testing method, this method can save the economic cost and time cost of parameter determination.

[0114] Furthermore, the gas diffusion layer parameter determination device provided in the embodiment of the present application can quickly revise the parameters through a parameter revision model when the simulation fails, which can further improve the efficiency of determining the gas diffusion layer parameters.

[0115] The device for determining parameters of a gas diffusion layer provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here.

[0116] The specific definition of the parameter determination device of the gas diffusion layer can refer to the definition of the parameter determination method of the gas diffusion layer above, which will not be repeated here. Each module in the above-mentioned parameter determination device of the gas diffusion layer can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the server in the form of hardware, or can be stored in the memory in the server in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0117] In another embodiment of the present application, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of the method for determining parameters of the gas diffusion layer of the embodiment of the present application are implemented.

[0118] In another embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for determining parameters of a gas diffusion layer in the embodiment of the present application are implemented.

[0119] In another embodiment of the present application, a computer program product is also provided. The computer program product includes computer instructions. When the computer instructions are executed on a parameter determination device of a gas diffusion layer, the parameter determination device of the gas diffusion layer executes each step of the gas diffusion layer parameter determination method in the method flow shown in the above method embodiment.

[0120] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loading and executing a computer execution instruction on a computer, a process or function according to an embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, data center, etc. that contains one or more servers that can be integrated with a medium. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0121] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0122] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for determining parameters of a gas diffusion layer, characterized in that: The method comprises: Acquire a simulation model of a gas diffusion layer, wherein the simulation model includes simulation parameters; Obtaining parameter values ​​corresponding to the simulation parameters; Performing simulation calculation on the simulation model and parameter values ​​corresponding to the simulation parameters to obtain a first simulation result; If the first simulation result is used to indicate a simulation failure, inputting prompt information in the first simulation result into a trained parameter revision model to obtain a revised parameter value, wherein the prompt information includes: a simulation failure type and setting parameter information, and the parameter revision model is a model based on a deep Q network; Performing simulation calculation on the simulation model and the revised parameter value to obtain a second simulation result; If the second simulation result is used to indicate that the simulation is successful, determining the revised parameter value as the target parameter value; Before obtaining the simulation model of the gas diffusion layer, the method further includes: Obtain different simulation models of multiple gas diffusion layers and prompt information of each simulation model; The parameter revision model is trained according to different simulation models and prompt information of each simulation model to obtain the trained parameter revision model.

2. The method according to claim 1, characterized in that The obtaining of the parameter value corresponding to the simulation parameter includes: The simulation model is input into the trained first model to obtain a value range of a parameter value corresponding to each simulation parameter, and the parameter value of each simulation parameter is determined from the value range to obtain the parameter value corresponding to the simulation parameter.

3. The method according to claim 1, characterized in that: The second simulation result includes: a compression ratio of the gas diffusion layer and a displacement corresponding to the compression ratio; If the second simulation result is used to indicate that the simulation is successful, determining the revised parameter value as the target parameter value includes: If the second simulation result is used to indicate that the simulation is successful, then when it is determined that the displacement corresponding to the compression ratio in the second simulation result is within a preset range, the revised parameter value is determined as the target parameter value.

4. The method according to claim 1, characterized in that: The method further comprises: Obtaining the compression ratio of the simulation parameter under different target parameter values, and the displacement corresponding to each compression ratio; According to the displacement change under different compression ratios, the sensitive parameters in the simulation parameters are determined.

5. The method according to claim 1, characterized in that The simulation parameters include a simulation step size and a calculation domain, and the method further includes: Determining the calculation time of the simulation model according to the simulation step size and the calculation domain; The performing simulation calculation on the simulation model and the parameter values ​​corresponding to the simulation parameters includes: If the calculation time is less than a preset threshold, simulation calculation is performed on the simulation model and the parameter values ​​corresponding to the simulation parameters.

6. The method according to claim 2, characterized in that Before inputting the simulation model into the trained first model, the method further comprises: Acquire different simulation models of multiple gas diffusion layers and target parameter values ​​corresponding to the simulation models; The parameter revision model is trained according to different simulation models and target parameter values ​​corresponding to each simulation model to obtain the trained first model.

7. A device for determining parameters of a gas diffusion layer, characterized in that: The device comprises: An acquisition module, used for acquiring a simulation model of a gas diffusion layer, wherein the simulation model includes simulation parameters and acquiring parameter values ​​corresponding to the simulation parameters; A first simulation module, used for performing simulation calculation on the simulation model and the parameter values ​​corresponding to the simulation parameters to obtain a first simulation result; A processing module, configured to input prompt information in the first simulation result into a trained parameter revision model to obtain a revised parameter value if the simulation result indicates a simulation failure, wherein the prompt information includes: a simulation failure type and parameter setting information, and the parameter revision model is a model based on a deep Q network; A second simulation module, used for performing simulation calculation on the simulation model and the revised parameter value to obtain a second simulation result; a determination module, configured to determine the revised parameter value as a target parameter value if the second simulation result indicates that the simulation is successful; A training module, used to obtain different simulation models of multiple gas diffusion layers and prompt information of each simulation model; The parameter revision model is trained according to different simulation models and prompt information of each simulation model to obtain the trained parameter revision model.

8. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method for determining parameters of the gas diffusion layer according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the method for determining parameters of a gas diffusion layer according to any one of claims 1 to 6 is implemented.

Images