Bipolar plate runner design method and system, storage medium and program product

Abstract

The invention discloses a bipolar plate flow channel design method and system, a storage medium and a program product, and relates to the technical field of fuel cells, the method comprises the following steps: simulating a sample bipolar plate flow channel by using a preset simulation model to obtain a simulation data set; according to the simulation data set and a preset graph neural network model, training to obtain a bipolar plate flow channel optimization model; based on a preset judgment basis and the size of the bipolar plate flow channel to be optimized, the bipolar plate flow channel optimization model is utilized to determine bipolar plate flow channel optimization parameters, and the preset judgment basis comprises preset gas concentration uniformity, preset pressure drop, preset liquid water content and preset current density uniformity; and determining structural parameters of the bipolar plate flow channel according to the bipolar plate flow channel optimization parameters. According to the method, the technical problem that CFD simulation lacks clear guidance for design of the flow channel type, the flow channel size, the depth and the ridge size of the bipolar plate is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of fuel cells, and in particular, relates to a bipolar plate flow channel design method, system, storage medium and program product. Background Art

[0002] A hydrogen fuel cell is a power generation device that directly converts the chemical energy of hydrogen and oxygen into electrical energy. As a key component in a fuel cell stack, the performance of a hydrogen fuel cell bipolar plate directly affects the performance and life of the fuel cell system. The flow channel design of the bipolar plate has an important influence on the uniform distribution of gas, the efficient contact between gas and catalyst layer, the transmission of liquid, the heat exchange between gas and liquid, etc. Therefore, the flow channel design is the key to the design of bipolar plates. In order to obtain a better flow channel design, the existing design methods usually adopt numerical analysis methods based on CFD (Computational Fluid Dynamics) or experimental methods. The CFD method provides a basis for design optimization by simulating and analyzing the airflow, pressure distribution and liquid flow of the flow channel; the experimental method obtains the performance of the flow channel by experimentally measuring the distribution of gas, water, etc. in the flow channel, and optimizes the flow channel design based on the experimental data.

[0003] With the development of computer technology, using CFD technology to analyze the designed bipolar plate flow channels has become a method adopted by most researchers. However, CFD simulation can only simulate and analyze the gas distribution, pressure loss, water content, etc. in the flow channels, and lacks clear guidance on the design of bipolar plate flow channel type, flow channel size, depth, and ridge size. Summary of the invention

[0004] Embodiments of the present invention provide a bipolar plate flow channel design method, system, storage medium and program product to at least solve the technical problem that CFD simulation lacks clear guidance on the bipolar plate flow channel type, flow channel size, depth and ridge size design.

[0005] According to a first aspect of an embodiment of the present invention, a bipolar plate flow channel design method is provided, comprising: using a preset simulation model to simulate a sample bipolar plate flow channel to obtain a simulation data set; training a bipolar plate flow channel optimization model based on the simulation data set and a preset graph neural network model; based on preset judgment criteria and the bipolar plate flow channel size to be optimized, using the bipolar plate flow channel optimization model, determining the bipolar plate flow channel optimization parameters, wherein the preset judgment criteria include a preset gas concentration uniformity, a preset pressure drop, a preset liquid water content, and a preset current density uniformity; and determining the bipolar plate flow channel structural parameters based on the bipolar plate flow channel optimization parameters.

[0006] Optionally, a preset simulation model is used to simulate the sample bipolar plate flow channels to obtain a simulation data set, including: using the preset simulation model to simulate multiple sample bipolar plate flow channels of different sizes to obtain data to be verified; verifying the data to be verified based on the sensor data to obtain a verification result, wherein the sensor data is data of multiple sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; based on the verification result, the preset simulation model is optimized to obtain an optimized simulation model; based on the optimized simulation model, multiple sample bipolar plate flow channels of different sizes are simulated to obtain a simulation data set.

[0007] Optionally, the preset simulation model is constructed based on computational fluid dynamics according to preset simulation requirements.

[0008] Optionally, a bipolar plate flow channel optimization model is trained based on a simulation data set and a preset graph neural network model, including: preprocessing the simulation data set to obtain a standard data set; training the preset graph neural network model based on the standard data set to obtain the bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

[0009] Optionally, the bipolar plate flow channel structural parameters are determined based on the bipolar plate flow channel optimization parameters, including: using sensor data to verify the bipolar plate flow channel optimization parameters to obtain a verification result, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; in response to the verification result indicating that the verification is passed, the bipolar plate flow channel structural parameters are determined.

[0010] Optionally, the bipolar plate channel optimization parameters include: gas concentration, pressure, water vapor content, liquid water content, current density, and the bipolar plate channel structure parameters include: number of grooves, groove width, groove depth and ridge width, ridge depth.

[0011] According to a second aspect of an embodiment of the present invention, a bipolar plate flow channel design system is also provided, including: a simulation module, used to simulate a sample bipolar plate flow channel using a preset simulation model to obtain a simulation data set; a training module, used to train a bipolar plate flow channel optimization model based on the simulation data set and a preset graph neural network model; a prediction module, used to determine the bipolar plate flow channel optimization parameters based on a preset judgment basis and the bipolar plate flow channel size to be optimized using the bipolar plate flow channel optimization model, wherein the preset judgment basis includes a preset gas concentration uniformity, a preset pressure drop, a preset liquid water content, and a preset current density uniformity; a determination module, used to determine the bipolar plate flow channel structural parameters based on the bipolar plate flow channel optimization parameters.

[0012] Optionally, the simulation module is also used to: use a preset simulation model to simulate multiple sample bipolar plate flow channels of different sizes to obtain data to be verified; verify the data to be verified based on the sensor data to obtain a verification result, wherein the sensor data is data of multiple sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; optimize the preset simulation model based on the verification result to obtain an optimized simulation model; simulate multiple sample bipolar plate flow channels of different sizes based on the optimized simulation model to obtain a simulation data set.

[0013] Optionally, the preset simulation model used by the simulation module is constructed based on computational fluid dynamics according to preset simulation requirements.

[0014] Optionally, the training module is also used to: preprocess the simulation data set to obtain a standard data set; train a preset graph neural network model based on the standard data set to obtain a bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

[0015] Optionally, the determination module is also used to: use sensor data to verify the bipolar plate flow channel optimization parameters to obtain a verification result, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; in response to the verification result indicating that the verification is passed, determine the bipolar plate flow channel structure parameters.

[0016] Optionally, the bipolar plate flow channel optimization parameters in the determination module include: gas concentration, pressure, water vapor content, liquid water content, current density, and the bipolar plate flow channel structural parameters in the determination module include: groove number, groove width, groove depth and ridge width, ridge depth.

[0017] According to a third aspect of an embodiment of the present invention, there is also provided an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the bipolar plate flow channel design method described in any embodiment of the first aspect above.

[0018] According to a fourth aspect of an embodiment of the present invention, a non-volatile storage medium is also provided, in which a computer program is stored, wherein the computer program is configured to execute the bipolar plate flow channel design method described in any embodiment of the first aspect when running on a computer or processor.

[0019] According to a fifth aspect of the embodiments of the present invention, a computer program product is also provided, including a computer program, which, when executed by a processor, implements the bipolar plate flow channel design method described in any embodiment of the first aspect above.

[0020] In an embodiment of the present invention, a preset simulation model is used to simulate the sample bipolar plate flow channel to obtain a simulation data set; a bipolar plate flow channel optimization model is trained based on the simulation data set and the preset graph neural network model; based on the preset judgment basis and the bipolar plate flow channel size to be optimized, the bipolar plate flow channel optimization model is used to determine the bipolar plate flow channel optimization parameters, wherein the preset judgment basis includes a preset gas concentration uniformity, a preset pressure drop, a preset liquid water content, and a preset current density uniformity; and the bipolar plate flow channel structural parameters are determined based on the bipolar plate flow channel optimization parameters. The present invention utilizes the constructed bipolar plate flow channel optimization model to further determine the bipolar plate flow channel structural parameters after determining the bipolar plate flow channel optimization parameters. The bipolar plate flow channel structural parameters can guide the bipolar plate flow channel type, flow channel size, depth, and ridge size design, thereby solving the technical problem that CFD simulation lacks clear guidance on the bipolar plate flow channel type, flow channel size, depth, and ridge size design. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0022] Figure 1 is a flow chart of a bipolar plate flow channel design method according to one embodiment of the present invention;

[0023] Figure 2 4 is a structural block diagram of a bipolar plate flow channel design system according to one embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] According to an embodiment of the present invention, an embodiment of a bipolar plate flow channel design method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system comprising at least one set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] The method embodiment can also be executed in an electronic device including a memory and a processor, a similar control device or a cloud. Taking an electronic device as an example, the electronic device may include one or more processors and a memory for storing data. Optionally, the electronic device may also include a communication device and a display device for communication functions. It can be understood by those of ordinary skill in the art that the above structural description is only illustrative and does not limit the structure of the above electronic device. For example, the electronic device may also include more or fewer components than the above structural description, or have a configuration different from the above structural description.

[0028] The processor may include one or more processing units. For example, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processing (DSP) chip, a microcontroller unit (MCU), a field-programmable gate array (FPGA), a neural network processor (NPU), a tensor processing unit (TPU), an artificial intelligence (AI) type processor, and the like. Among them, different processing units may be independent components or integrated into one or more processors. In some instances, the electronic device may also include one or more processors.

[0029] The memory can be used to store computer programs, such as storing a computer program corresponding to the bipolar plate flow channel design method in an embodiment of the present invention, and the processor implements the above-mentioned bipolar plate flow channel design method by running the computer program stored in the memory. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0030] The communication device is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the communication device includes a network adapter (network interface controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the communication device can be a radio frequency (radio frequency, RF) module, which is used to communicate with the Internet wirelessly. In some embodiments of the present solution, the communication device is used to connect to mobile devices such as mobile phones and tablets, and instructions can be sent to electronic devices through the mobile device.

[0031] The display device may be a touch screen type liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or "touch display screen"). The liquid crystal display may enable a user to interact with a user interface of the electronic device. In some embodiments, the electronic device has a graphical user interface (GUI), and a user may interact with the GUI by touching a finger release and / or gestures on a touch-sensitive surface, and executable instructions for executing the above-mentioned human-computer interaction functions are configured / stored in a computer program product or a readable storage medium executable by one or more processors.

[0032] Figure 1 FIG. 1 is a flow chart of a bipolar plate flow channel design method according to one embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0033] Step S101, using a preset simulation model, simulate the flow channel of a sample bipolar plate to obtain a simulation data set.

[0034] Step S102, training a bipolar plate flow channel optimization model based on the simulation data set and the preset graph neural network model.

[0035] Step S103, based on preset judgment criteria and the bipolar plate flow channel size to be optimized, using the bipolar plate flow channel optimization model, determine the bipolar plate flow channel optimization parameters, wherein the preset judgment criteria include preset gas concentration uniformity, preset pressure drop, preset liquid water content, and preset current density uniformity.

[0036] Specifically, after the preset judgment basis and the bipolar plate flow channel size to be optimized are input into the bipolar plate flow channel optimization model, the bipolar plate flow channel optimization model can determine the parameter distribution of the bipolar plate flow channel corresponding to the bipolar plate flow channel size to be optimized under different flow channel shapes according to the input data, and then determine the bipolar plate flow channel optimization parameters according to the preset judgment basis and parameter distribution. Among them, the parameters in the parameter distribution refer to gas concentration distribution, pressure drop, water vapor content, liquid water content and current density.

[0037] Step S104, determining bipolar plate flow channel structural parameters according to bipolar plate flow channel optimization parameters.

[0038] Specifically, in the process of determining the structural parameters of the bipolar plate flow channels according to the bipolar plate flow channel optimization parameters, the structural parameters of the bipolar plate flow channels are continuously adjusted until the gas concentration, pressure, water vapor content, liquid water content, and current density of the corresponding bipolar plate flow channels can meet the bipolar plate flow channel optimization parameters.

[0039] In an embodiment of the present invention, a preset simulation model is used to simulate the sample bipolar plate flow channel to obtain a simulation data set; a bipolar plate flow channel optimization model is trained based on the simulation data set and the preset graph neural network model; based on the preset judgment basis and the bipolar plate flow channel size to be optimized, the bipolar plate flow channel optimization model is used to determine the bipolar plate flow channel optimization parameters, wherein the preset judgment basis includes a preset gas concentration uniformity, a preset pressure drop, a preset liquid water content, and a preset current density uniformity; and the bipolar plate flow channel structural parameters are determined based on the bipolar plate flow channel optimization parameters. The present invention utilizes the constructed bipolar plate flow channel optimization model to further determine the bipolar plate flow channel structural parameters after determining the bipolar plate flow channel optimization parameters. The bipolar plate flow channel structural parameters can guide the bipolar plate flow channel type, flow channel size, depth, and ridge size design, thereby solving the technical problem that CFD simulation lacks clear guidance on the bipolar plate flow channel type, flow channel size, depth, and ridge size design.

[0040] Optionally, a preset simulation model is used to simulate the sample bipolar plate flow channels to obtain a simulation data set, including: using the preset simulation model to simulate multiple sample bipolar plate flow channels of different sizes to obtain data to be verified; verifying the data to be verified based on the sensor data to obtain a verification result, wherein the sensor data is data of multiple sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; based on the verification result, the preset simulation model is optimized to obtain an optimized simulation model; based on the optimized simulation model, multiple sample bipolar plate flow channels of different sizes are simulated to obtain a simulation data set.

[0041] Optionally, the preset simulation model is constructed based on computational fluid dynamics according to preset simulation requirements.

[0042] Optionally, a bipolar plate flow channel optimization model is trained based on a simulation data set and a preset graph neural network model, including: preprocessing the simulation data set to obtain a standard data set; training the preset graph neural network model based on the standard data set to obtain the bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

[0043] Optionally, the bipolar plate flow channel structural parameters are determined based on the bipolar plate flow channel optimization parameters, including: using sensor data to verify the bipolar plate flow channel optimization parameters to obtain a verification result, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; in response to the verification result indicating that the verification is passed, the bipolar plate flow channel structural parameters are determined.

[0044] Optionally, the bipolar plate channel optimization parameters include: gas concentration, pressure, water vapor content, liquid water content, current density, and the bipolar plate channel structure parameters include: number of grooves, groove width, groove depth and ridge width, ridge depth.

[0045] Specifically, based on the above embodiments, the example implementation of the present invention is as follows:

[0046] (1) CFD model construction and database establishment:

[0047] This patent uses a numerical simulation method (CFD) to simulate the bipolar plate flow channels of different hydrogen fuel cell stacks, obtains simulation data of different sizes under the same flow channel, verifies the corresponding data through sensors, corrects the corresponding data and model, and then applies the CFD to other simulations.

[0048] (2) Construction of graph neural network model:

[0049] 1) Construction of monitoring point network:

[0050] Monitoring point definition: Each monitoring point is regarded as a node, and its specific features include position coordinates, gas concentration, pressure, water vapor content, liquid water content, and current density. The position coordinates of node i can be expressed as (xi, yi), and the coordinates of each node are collected to form a feature vector.

[0051] Edge definition: The connection between nodes is determined by the physical distance and direction vector between sensors, which is defined as a fully connected node, that is, all nodes are connected to each other, and the edge is represented by E ij =(v ij ,d ij ), where v ij represents the position vector of node i pointing to node j, d ij Indicates the node distance.

[0052] 2) Data preprocessing of graph structure:

[0053] Normalization: All sensor data are normalized to eliminate the impact of dimension and unify the data scale. Common methods include minimum-maximum normalization or Z-score normalization:

[0054]

[0055] Here, μ and σ are the mean and standard deviation respectively.

[0056] 3. Architecture of the GCN model:

[0057] (1) Input layer: Receives the input data of the model and builds a monitoring point network based on the input data of the model.

[0058] (2) Graph convolutional layer: It contains several (2-5) graph convolutional layers, each of which uses the adjacent node information to update and share features:

[0059] (3) Output layer: The final output model predicts parameters such as gas concentration, gas pressure, water vapor content, liquid water content, and current density. The bipolar plate flow channel structure is optimized based on the gas concentration, gas pressure, water vapor content, liquid water content, and current density to obtain a bipolar plate flow channel with good performance.

[0060] 4. Training and validation:

[0061] (1) Loss function: Use the binary cross entropy loss function to optimize the classification task:

[0062]

[0063] Among them, y ij is the true label, is the predicted value, N is the number of samples, and C is the number of categories.

[0064] (2) Optimizer: Use the Adam optimizer to adjust model parameters to improve training efficiency.

[0065] (3) Validation: Evaluate the model’s test performance on the test set through holdout or cross-validation.

[0066] 3) Bipolar plate optimization design system combined with graph convolutional neural network:

[0067] 1. Data collection and processing:

[0068] Sensor data collection: Verify the results of the graph convolutional neural network and design experiments for further verification.

[0069] 2. Optimization design of flow channel GCN model:

[0070] Model reasoning: The preprocessed sensor data is fed into the GCN model for real-time prediction. The model uses current and historical data to estimate the various parameters of each area (gas concentration, pressure, water vapor content, liquid water content, current density, etc.), thereby optimizing the design of the bipolar plate flow channel structure parameters.

[0071] 3. System maintenance and iterative updates:

[0072] Data feedback loop: The data collected during the system operation will be used to further train and optimize the GCN model to improve the accuracy of warnings and response speed.

[0073] Model optimization: GCN models are regularly evaluated and adjusted to adapt to environmental changes and technology upgrades.

[0074] It should be noted that the graph convolutional layer (GCL) is a special convolutional layer that is applied to graph structured data rather than traditional image data. In the graph convolutional network (GCN), the graph convolutional layer learns the representation of nodes by aggregating and propagating the nodes on the graph. The following is a detailed description of how the graph convolutional layer processes the input feature vector:

[0075] 1. Input feature vector: The input of the graph convolution layer is a set of node feature vectors. These vectors can be the features of the original data or the features processed by the previous layer.

[0076] 2. Aggregation operation: In the graph convolution layer, the feature vector of each node is aggregated with the feature vectors of its neighboring nodes. This aggregation process can be performed in many ways, and common methods include average aggregation and weighted aggregation.

[0077] 3. Propagation operation: After aggregation, the aggregated information needs to be propagated back to each node. This process usually involves a weight matrix, which defines how to propagate the weighted sum of the information of neighboring nodes back.

[0078] 4. Non-linear activation function: After propagation, a non-linear activation function, such as ReLU, is usually applied to increase the expressive power of the model.

[0079] 5. Output feature vector: Ultimately, the output feature vector of each node is determined by its hidden state, which can be used as the input of the next layer or as the final output of the graph convolutional network.

[0080] It should be noted that using deep learning to optimize the shape and size of hydrogen fuel cell flow channels is a multi-faceted method. The following are some optional methods and steps:

[0081] Data collection: First, a large amount of hydrogen fuel cell flow channel design data needs to be collected, including successful and failed designs. This data can come from public academic papers, patent documents, experimental data, etc. This data can be organized into a standardized format for training deep learning models.

[0082] Feature extraction: Extract useful features from the collected data, which can include the geometry and size of the flow channel.

[0083] Model selection: Select a deep learning model suitable for shape optimization.

[0084] Loss function design: Design a suitable loss function to measure the quality of hydrogen fuel cell flow channel design. This loss function should be able to reflect the performance of the flow channel in terms of gas transmission, liquid water discharge, current density, stability, etc. It may be necessary to combine multiple factors, such as the length, width, depth, material properties, etc. of the flow channel.

[0085] Model training: Use the collected data to train the deep learning model. This may require a lot of computing resources and a long training process. During the training process, the back-propagation algorithm can be used to optimize the weights of the model.

[0086] Evaluation and iteration: Use some criteria to evaluate the performance of the model, such as accuracy, recall, F1 score, etc. If the model's performance is poor, you may need to go back to the previous steps and redesign the feature extraction method, loss function, or model structure. This process may require multiple iterations until satisfactory performance is achieved.

[0087] Generate new designs: Use the trained model to generate new hydrogen fuel cell flow path designs. These designs can serve as the basis for subsequent manual design and testing.

[0088] Verification and testing: Finally, the generated new designs need to be rigorously verified and tested to ensure their performance and safety in actual applications. This may include experimental testing, simulation calculations, actual application testing, etc.

[0089] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0090] In this embodiment, a bipolar plate flow channel design system is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions that have been made will not be repeated. As used below, the term "module" is a combination of software and / or hardware that can implement a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0091] Figure 2 is a structural block diagram of a bipolar plate flow channel design system 200 according to one embodiment of the present invention. Figure 2As shown, a bipolar plate flow channel design system 200 is taken as an example, including: a simulation module 201, which is used to simulate the sample bipolar plate flow channel using a preset simulation model to obtain a simulation data set; a training module 202, which is used to train a bipolar plate flow channel optimization model based on the simulation data set and a preset graph neural network model; a prediction module 203, which is used to determine the bipolar plate flow channel optimization parameters based on the preset judgment basis and the bipolar plate flow channel size to be optimized using the bipolar plate flow channel optimization model, wherein the preset judgment basis includes a preset gas concentration uniformity, a preset pressure drop, a preset liquid water content, and a preset current density uniformity; a determination module 204, which is used to determine the bipolar plate flow channel structural parameters according to the bipolar plate flow channel optimization parameters.

[0092] Optionally, the simulation module 201 is also used to: use a preset simulation model to simulate multiple sample bipolar plate flow channels of different sizes to obtain data to be verified; verify the data to be verified based on the sensor data to obtain a verification result, wherein the sensor data is data of multiple sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; optimize the preset simulation model based on the verification result to obtain an optimized simulation model; simulate multiple sample bipolar plate flow channels of different sizes based on the optimized simulation model to obtain a simulation data set.

[0093] Optionally, the preset simulation model used by the simulation module 201 is constructed based on computational fluid dynamics according to preset simulation requirements.

[0094] Optionally, the training module 202 is also used to: preprocess the simulation data set to obtain a standard data set; train a preset graph neural network model based on the standard data set to obtain a bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

[0095] Optionally, the determination module 204 is also used to: use sensor data to verify the bipolar plate flow channel optimization parameters to obtain a verification result, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; in response to the verification result indicating that the verification is passed, determine the bipolar plate flow channel structure parameters.

[0096] Optionally, the bipolar plate channel optimization parameters in the determination module 204 include: gas concentration, pressure, water vapor content, liquid water content, and current density; the bipolar plate channel structure parameters in the determination module 204 include: number of grooves, groove width, groove depth and ridge width, ridge depth.

[0097] An embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the bipolar plate flow channel design method described in any of the above embodiments.

[0098] Optionally, in this embodiment, the processor in the above electronic device may be configured to run a computer program to perform the following steps:

[0099] Step S101, using a preset simulation model, simulate the flow channel of a sample bipolar plate to obtain a simulation data set.

[0100] Step S102, training a bipolar plate flow channel optimization model based on the simulation data set and the preset graph neural network model.

[0101] Step S103, based on preset judgment criteria and the bipolar plate flow channel size to be optimized, using the bipolar plate flow channel optimization model, determine the bipolar plate flow channel optimization parameters, wherein the preset judgment criteria include preset gas concentration uniformity, preset pressure drop, preset liquid water content, and preset current density uniformity.

[0102] Step S104, determining bipolar plate flow channel structural parameters according to bipolar plate flow channel optimization parameters.

[0103] Optionally, a preset simulation model is used to simulate the sample bipolar plate flow channels to obtain a simulation data set, including: using the preset simulation model to simulate multiple sample bipolar plate flow channels of different sizes to obtain data to be verified; verifying the data to be verified based on the sensor data to obtain a verification result, wherein the sensor data is data of multiple sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; based on the verification result, the preset simulation model is optimized to obtain an optimized simulation model; based on the optimized simulation model, multiple sample bipolar plate flow channels of different sizes are simulated to obtain a simulation data set.

[0104] Optionally, the preset simulation model is constructed based on computational fluid dynamics according to preset simulation requirements.

[0105] Optionally, a bipolar plate flow channel optimization model is trained based on a simulation data set and a preset graph neural network model, including: preprocessing the simulation data set to obtain a standard data set; training the preset graph neural network model based on the standard data set to obtain the bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

[0106] Optionally, the bipolar plate flow channel structural parameters are determined based on the bipolar plate flow channel optimization parameters, including: using sensor data to verify the bipolar plate flow channel optimization parameters to obtain a verification result, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; in response to the verification result indicating that the verification is passed, the bipolar plate flow channel structural parameters are determined.

[0107] Optionally, the bipolar plate channel optimization parameters include: gas concentration, pressure, water vapor content, liquid water content, current density, and the bipolar plate channel structure parameters include: number of grooves, groove width, groove depth and ridge width, ridge depth.

[0108] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0109] An embodiment of the present invention further provides a non-volatile storage medium, in which a computer program is stored, wherein the computer program is configured to execute the bipolar plate flow channel design method described in any of the above embodiments when running on a computer or a processor.

[0110] Optionally, in this embodiment, the above-mentioned computer program may be configured to store a computer program for performing the following steps:

[0111] Step S101, using a preset simulation model, simulate the flow channel of a sample bipolar plate to obtain a simulation data set.

[0112] Step S102, training a bipolar plate flow channel optimization model based on the simulation data set and the preset graph neural network model.

[0113] Step S103, based on preset judgment criteria and the bipolar plate flow channel size to be optimized, using the bipolar plate flow channel optimization model, determine the bipolar plate flow channel optimization parameters, wherein the preset judgment criteria include preset gas concentration uniformity, preset pressure drop, preset liquid water content, and preset current density uniformity.

[0114] Step S104, determining bipolar plate flow channel structural parameters according to bipolar plate flow channel optimization parameters.

[0115] Optionally, a preset simulation model is used to simulate the sample bipolar plate flow channels to obtain a simulation data set, including: using the preset simulation model to simulate multiple sample bipolar plate flow channels of different sizes to obtain data to be verified; verifying the data to be verified based on the sensor data to obtain a verification result, wherein the sensor data is data of multiple sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; based on the verification result, the preset simulation model is optimized to obtain an optimized simulation model; based on the optimized simulation model, multiple sample bipolar plate flow channels of different sizes are simulated to obtain a simulation data set.

[0116] Optionally, the preset simulation model is constructed based on computational fluid dynamics according to preset simulation requirements.

[0117] Optionally, a bipolar plate flow channel optimization model is trained based on a simulation data set and a preset graph neural network model, including: preprocessing the simulation data set to obtain a standard data set; training the preset graph neural network model based on the standard data set to obtain the bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

[0118] Optionally, the bipolar plate flow channel structural parameters are determined based on the bipolar plate flow channel optimization parameters, including: using sensor data to verify the bipolar plate flow channel optimization parameters to obtain a verification result, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; in response to the verification result indicating that the verification is passed, the bipolar plate flow channel structural parameters are determined.

[0119] Optionally, the bipolar plate channel optimization parameters include: gas concentration, pressure, water vapor content, liquid water content, current density, and the bipolar plate channel structure parameters include: number of grooves, groove width, groove depth and ridge width, ridge depth.

[0120] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0121] An embodiment of the present invention further provides a computer program product, including a computer program, which implements the bipolar plate flow channel design method described in any of the above embodiments when executed by a processor.

[0122] Optionally, in this embodiment, the above computer program implements the following steps when executed by a processor:

[0123] Step S101, using a preset simulation model, simulate the flow channel of a sample bipolar plate to obtain a simulation data set.

[0124] Step S102, training a bipolar plate flow channel optimization model based on the simulation data set and the preset graph neural network model.

[0125] Step S103, based on preset judgment criteria and the bipolar plate flow channel size to be optimized, using the bipolar plate flow channel optimization model, determine the bipolar plate flow channel optimization parameters, wherein the preset judgment criteria include preset gas concentration uniformity, preset pressure drop, preset liquid water content, and preset current density uniformity.

[0126] Step S104, determining bipolar plate flow channel structural parameters according to bipolar plate flow channel optimization parameters.

[0127] Optionally, a preset simulation model is used to simulate the sample bipolar plate flow channels to obtain a simulation data set, including: using the preset simulation model to simulate multiple sample bipolar plate flow channels of different sizes to obtain data to be verified; verifying the data to be verified based on the sensor data to obtain a verification result, wherein the sensor data is data of multiple sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; based on the verification result, the preset simulation model is optimized to obtain an optimized simulation model; based on the optimized simulation model, multiple sample bipolar plate flow channels of different sizes are simulated to obtain a simulation data set.

[0128] Optionally, the preset simulation model is constructed based on computational fluid dynamics according to preset simulation requirements.

[0129] Optionally, a bipolar plate flow channel optimization model is trained based on a simulation data set and a preset graph neural network model, including: preprocessing the simulation data set to obtain a standard data set; training the preset graph neural network model based on the standard data set to obtain the bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

[0130] Optionally, the bipolar plate flow channel structural parameters are determined based on the bipolar plate flow channel optimization parameters, including: using sensor data to verify the bipolar plate flow channel optimization parameters to obtain a verification result, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, current density; in response to the verification result indicating that the verification is passed, the bipolar plate flow channel structural parameters are determined.

[0131] Optionally, the bipolar plate channel optimization parameters include: gas concentration, pressure, water vapor content, liquid water content, current density, and the bipolar plate channel structure parameters include: number of grooves, groove width, groove depth and ridge width, ridge depth.

[0132] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0133] In some embodiments provided in the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the modules can be a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, modules or indirect coupling or communication connection of modules, which can be electrical or other forms.

[0134] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0135] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0136] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0137] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A bipolar plate flow channel design method, characterized in that: include: Using a preset simulation model, the flow channel of the sample bipolar plate is simulated to obtain a simulation data set; According to the simulation data set and the preset graph neural network model, a bipolar plate flow channel optimization model is trained; Based on the preset judgment basis and the bipolar plate flow channel size to be optimized, the bipolar plate flow channel optimization model is used to determine the bipolar plate flow channel optimization parameters, wherein the preset judgment basis includes a preset gas concentration uniformity, a preset pressure drop, a preset liquid water content, and a preset current density uniformity; According to the bipolar plate flow channel optimization parameters, the bipolar plate flow channel structural parameters are determined.

2. The bipolar plate flow channel design method according to claim 1, characterized in that: The preset simulation model is used to simulate the flow channel of the sample bipolar plate to obtain a simulation data set, including: Using the preset simulation model, simulate a plurality of sample bipolar plate flow channels of different sizes to obtain data to be verified; According to the sensor data, the data to be verified is verified to obtain a verification result, wherein the sensor data is data of the plurality of sample bipolar plate flow channels of different sizes obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, and current density; According to the verification result, optimizing the preset simulation model to obtain an optimized simulation model; According to the optimized simulation model, the sample bipolar plate flow channels of the plurality of different sizes are simulated to obtain the simulation data set.

3. The bipolar plate flow channel design method according to claim 1, characterized in that: The preset simulation model is constructed based on computational fluid dynamics according to preset simulation requirements.

4. The bipolar plate flow channel design method according to claim 1, characterized in that: According to the simulation data set and the preset graph neural network model, a bipolar plate flow channel optimization model is trained, including: Preprocessing the simulation data set to obtain a standard data set; According to the standard data set, the preset graph neural network model is trained to obtain a bipolar plate flow channel optimization model, wherein the bipolar plate flow channel optimization model includes a monitoring point network, a graph convolutional network input layer, a graph convolutional layer, and a graph convolutional network output layer.

5. The bipolar plate flow channel design method according to claim 1, characterized in that: Determining bipolar plate flow channel structural parameters according to the bipolar plate flow channel optimization parameters includes: Using sensor data, verifying the bipolar plate flow channel optimization parameters to obtain verification results, wherein the sensor data is data of the bipolar plate flow channel to be optimized obtained by the sensor, including: gas concentration, pressure, water vapor content, liquid water content, and current density; In response to the verification result indicating that the verification is passed, the bipolar plate flow channel structure parameters are determined.

6. The bipolar plate flow channel design method according to claim 1, characterized in that: The bipolar plate channel optimization parameters include: gas concentration, pressure, water vapor content, liquid water content, and current density; the bipolar plate channel structure parameters include: channel number, channel width, channel depth, ridge width, and ridge depth.

7. A bipolar plate flow channel design system, characterized in that: include: A simulation module, used to simulate the flow channel of the sample bipolar plate using a preset simulation model to obtain a simulation data set; A training module, used for training a bipolar plate flow channel optimization model according to the simulation data set and a preset graph neural network model; A prediction module, for determining bipolar plate flow channel optimization parameters using the bipolar plate flow channel optimization model based on preset judgment criteria and the bipolar plate flow channel size to be optimized, wherein the preset judgment criteria include preset gas concentration uniformity, preset pressure drop, preset liquid water content, and preset current density uniformity; A determination module is used to determine the bipolar plate flow channel structure parameters according to the bipolar plate flow channel optimization parameters.

8. A non-volatile storage medium, characterized in that: The non-volatile storage medium stores a computer program, wherein the computer program is configured to execute the bipolar plate flow channel design method described in any one of claims 1 to 6 when running on a computer or a processor.

9. A computer program product, characterized in that It comprises a computer program, which, when executed by a processor, implements the bipolar plate flow channel design method according to any one of claims 1 to 6.

10. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to execute the bipolar plate flow channel design method described in any one of claims 1 to 6.

Images