Fuel cell membrane electrode thickness detection method, device, equipment and medium

Abstract

The invention discloses a fuel cell membrane electrode thickness detection method, device and equipment and a medium, and is applied to the technical field of fuel cells, and the method comprises the steps: obtaining the cross section thicknesses of a plurality of different positions of a target structure layer of a membrane electrode assembly, inputting the cross section thicknesses of the plurality of different positions of the target structure layer into a membrane electrode thickness detection model, the membrane electrode thickness detection model determines spatial domain thickness characteristic data and frequency domain thickness characteristic data of the target structure layer based on the cross section thicknesses of a plurality of different positions of the target structure layer; and analyzing the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain a thickness detection result used for representing the thickness uniformity and / or thickness attenuation degree of the target structure layer, so that the thickness uniformity and / or thickness attenuation degree can be detected by obtaining a limited number of cross section thicknesses. Therefore, the detection period and workload can be reduced, the labor and equipment detection cost is reduced, and the membrane electrode thickness detection accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of fuel cell technology, and in particular to a fuel cell membrane electrode thickness detection method, device, equipment and medium. Background Art

[0002] In a fuel cell system, the membrane electrode assembly is the basic unit for generating electricity. When the membrane electrode assembly is subjected to carbon corrosion, mechanical durability, anti-reverse polarity and other working condition tests of the small piece fixture, the thickness of the membrane electrode assembly may change compared to the original sample. The uniformity and attenuation of the thickness of the membrane electrode assembly are important indicators reflecting the degree of corrosion of the membrane electrode assembly. Therefore, thickness detection of the membrane electrode assembly is an important step before stacking the membrane electrode assembly. However, the thickness detection method for the membrane electrode assembly in the prior art often has a long cycle, a large workload, and high labor and equipment costs. Summary of the invention

[0003] The present application provides a fuel cell membrane electrode thickness detection method, device, equipment and medium to solve the problems of long cycle, large workload and high cost in the prior art for fuel cell membrane electrode thickness detection. Specifically, the technical solution provided by the present application is as follows:

[0004] On the one hand, the present application provides a fuel cell membrane electrode thickness detection method, comprising:

[0005] Obtaining cross-sectional thicknesses at multiple different locations of a target structural layer of a membrane electrode assembly;

[0006] The cross-sectional thicknesses at multiple different positions of the target structure layer are input into the membrane electrode thickness detection model to obtain the thickness detection results of the target structure layer output by the membrane electrode thickness detection model; wherein the membrane electrode thickness detection model is used to determine the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, and then analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain the thickness detection results used to characterize the thickness uniformity and / or thickness attenuation of the target structure layer.

[0007] In a possible implementation, obtaining the cross-sectional thickness of a target structural layer of a membrane electrode assembly at a plurality of different positions includes:

[0008] Collecting infrared sensing signals of cross sections at multiple different positions calibrated at set size intervals on the target structure layer through an infrared sensor, and determining the thickness of the cross sections at multiple different positions of the target structure layer based on the infrared sensing signals of the cross sections at multiple different positions of the target structure layer;

[0009] or;

[0010] A cross-sectional depth image and a cross-sectional RGB image of the target structure layer are collected by a binocular camera device, and image analysis is performed on the cross-sectional depth image and the cross-sectional RGB image to identify the cross-sectional thicknesses of a plurality of different positions calibrated at set size intervals on the target structure layer;

[0011] or;

[0012] The cross-sectional thickness of multiple different locations on the target structure layer calibrated at set size intervals is measured using a microscope.

[0013] In a possible implementation, based on the cross-sectional thicknesses at multiple different positions of the target structure layer, determining the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer includes:

[0014] Performing curve fitting on the cross-sectional thicknesses at multiple different positions of the target structure layer to obtain spatial domain thickness characteristic data for characterizing the nonlinear relationship between the plane position and the cross-sectional thickness of the target structure layer;

[0015] The spatial domain thickness characteristic data of the target structure layer is subjected to Fourier transformation to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak value.

[0016] In a possible implementation, Fourier transform is performed on the spatial domain thickness characteristic data of the target structure layer to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance and the cross-sectional thickness peak of the target structure layer, including:

[0017] Smoothing the spatial domain thickness characteristic data of the target structural layer to obtain smoothed spatial domain thickness characteristic data;

[0018] The smoothed spatial domain thickness characteristic data is subjected to Fourier transform to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak value.

[0019] In a possible implementation, the spatial domain thickness characteristic data and the frequency domain thickness characteristic data are analyzed to obtain a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation of the target structure layer, including:

[0020] Analyze the space domain thickness characteristic data and the frequency domain thickness characteristic data to obtain the periodic variation data of the cross-sectional thickness of the target structure layer;

[0021] Based on the periodic variation data of the cross-sectional thickness of the target structural layer, a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation of the target structural layer is determined.

[0022] In a possible implementation, the membrane electrode thickness detection model includes a plurality of convolutional layers for acquiring thickness feature data in the spatial domain and thickness feature data in the frequency domain, and a fully connected layer connected to the plurality of convolutional layers for acquiring thickness detection results.

[0023] In a possible implementation manner, before obtaining the cross-sectional thicknesses of a plurality of different positions of a target structural layer of a membrane electrode assembly, the method further includes:

[0024] Acquire a training sample set; wherein the training sample set includes a plurality of training sample data, each of which includes the cross-sectional thickness of a target structural layer of a membrane electrode assembly at a plurality of different positions before and after the operating condition test, as well as the standard thickness uniformity and standard thickness attenuation of the target structural layer;

[0025] The initial membrane electrode thickness detection model is trained based on the training sample set to obtain the membrane electrode thickness detection model.

[0026] On the other hand, the present application provides a fuel cell membrane electrode thickness detection device, comprising:

[0027] An acquisition unit, used for acquiring cross-sectional thicknesses at multiple different positions of a target structural layer of the membrane electrode assembly;

[0028] A detection unit is used to input the cross-sectional thicknesses at multiple different positions of the target structure layer into a membrane electrode thickness detection model to obtain a thickness detection result of the target structure layer output by the membrane electrode thickness detection model; wherein the membrane electrode thickness detection model is used to determine the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, and then analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation degree of the target structure layer.

[0029] In one possible embodiment, the acquisition unit is specifically used to collect cross-sectional infrared sensing signals at multiple different positions on the target structure layer calibrated according to set size intervals through an infrared sensor, and determine the cross-sectional thicknesses at multiple different positions of the target structure layer based on the cross-sectional infrared sensing signals at multiple different positions of the target structure layer; or; collect a cross-sectional depth image and a cross-sectional RGB image of the target structure layer through a binocular camera device, and perform image analysis on the cross-sectional depth image and the cross-sectional RGB image to identify the cross-sectional thicknesses at multiple different positions on the target structure layer calibrated according to set size intervals; or; measure the cross-sectional thicknesses at multiple different positions on the target structure layer calibrated according to set size intervals through a microscope.

[0030] In one possible embodiment, the detection unit is specifically used to perform curve fitting on the cross-sectional thicknesses at multiple different positions of the target structure layer to obtain spatial domain thickness characteristic data for characterizing the nonlinear relationship between the plane position and the cross-sectional thickness of the target structure layer; and perform Fourier transform on the spatial domain thickness characteristic data of the target structure layer to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak.

[0031] In a possible implementation, the detection unit is specifically used to smooth the spatial domain thickness characteristic data of the target structure layer to obtain smoothed spatial domain thickness characteristic data; and to perform Fourier transform on the smoothed spatial domain thickness characteristic data to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak value.

[0032] In one possible implementation, the detection unit is specifically used to analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain the periodic variation data of the cross-sectional thickness of the target structure layer; based on the periodic variation data of the cross-sectional thickness of the target structure layer, determine the thickness detection result used to characterize the thickness uniformity and / or thickness attenuation of the target structure layer.

[0033] In a possible implementation, the membrane electrode thickness detection model includes a plurality of convolutional layers for acquiring thickness feature data in the spatial domain and thickness feature data in the frequency domain, and a fully connected layer connected to the plurality of convolutional layers for acquiring thickness detection results.

[0034] In a possible implementation, the fuel cell membrane electrode thickness detection device provided in the present application further includes:

[0035] A training unit is used to obtain a training sample set, the training sample set includes multiple training sample data, each training sample data includes the cross-sectional thickness of a target structure layer of a membrane electrode assembly at multiple different positions before and after the operating condition test, as well as the standard thickness uniformity and standard thickness attenuation of the target structure layer; based on the training sample set, an initial membrane electrode thickness detection model is trained to obtain a membrane electrode thickness detection model.

[0036] On the other hand, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned fuel cell membrane electrode thickness detection method when executing the computer program.

[0037] On the other hand, the present application also provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the above-mentioned fuel cell membrane electrode thickness detection method is implemented.

[0038] The beneficial effects of this application are as follows:

[0039] The present application can use a membrane electrode thickness detection model to realize automatic detection of the thickness uniformity and / or thickness attenuation of the target structure layer by acquiring the cross-sectional thickness at multiple different positions of the target structure layer of the membrane electrode assembly, thereby realizing automatic detection of the thickness uniformity and / or thickness attenuation by acquiring a limited number of cross-sectional thicknesses, thereby reducing the cycle and workload of membrane electrode thickness detection, and reducing labor and equipment detection costs. Moreover, by using the membrane electrode thickness detection model to detect the thickness uniformity and / or thickness attenuation of the target structure layer based on the spatial domain thickness characteristic data and frequency domain thickness characteristic data of the target structure layer, the accuracy of the thickness uniformity and / or thickness attenuation of the membrane electrode can be improved.

[0040] Other features and advantages of the present application will be described in the subsequent description, and in part, will become apparent from the description, or will be understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0042] Figure 1 A schematic diagram of the composition structure of the membrane electrode assembly in the fuel cell system of this application;

[0043] Figure 2 The schematic diagram is a general flow chart of the fuel cell membrane electrode thickness detection method in this application;

[0044] Figure 3 Schematic diagram of the composition structure of the membrane electrode thickness detection model in this application;

[0045] Figure 4 A schematic diagram of a cross-sectional thickness curve of a target structural layer of a membrane electrode in this application;

[0046] Figure 5 A schematic diagram of a cross-sectional thickness spectrum diagram of a target structural layer of a membrane electrode in this application;

[0047] Figure 6 Schematic diagram of the training process of the membrane electrode thickness detection model in this application;

[0048] Figure 7 This is a functional structure diagram of the fuel cell membrane electrode thickness detection device in this application;

[0049] Figure 8 Schematic diagram of the hardware structure of the electronic device in this application. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and beneficial effects of this application clearer, the technical solution in the embodiment of this application will be clearly and completely described below in conjunction with the drawings in the embodiment of this application. Obviously, the described embodiment is only a part of the embodiment of this application, not all the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0051] In order to facilitate those skilled in the art to better understand the present application, the technical terms involved in the present application are briefly introduced below.

[0052] A single cell is a component of a fuel cell stack, consisting of a bipolar plate and a membrane electrode assembly (MEA). The membrane electrode assembly is the core component of a single cell power generation. Figure 1 As shown, it is composed of a gas diffusion layer, a catalyst layer and a proton exchange membrane, and specifically includes seven structural layers, namely, a proton exchange membrane, an anode catalyst layer and a cathode catalyst layer, an anode gas diffusion layer and a cathode gas diffusion layer, and a sealing gasket with sealing and fixing functions. The working principle of the membrane electrode assembly is that hydrogen undergoes an oxidation reaction on the surface of the anode catalyst layer under the action of the anode catalyst to generate hydrogen protons and electrons, the hydrogen protons pass through the proton exchange membrane, and undergo a reduction reaction with oxygen on the surface of the cathode catalyst layer under the action of the cathode catalyst to generate water, and the electrons are directionally conducted to the cathode catalyst layer through an external circuit to generate current in the process of directional conduction to provide electricity for the load.

[0053] The target structural layer is a structural layer to be tested that is configured according to actual needs among the various structural layers of the membrane electrode assembly, for example, the target structural layer is the entire membrane electrode assembly, or the target structural layer is the proton exchange membrane of the membrane electrode assembly, or the target structural layer is the proton exchange membrane, anode gas diffusion layer and cathode gas diffusion layer of the membrane electrode assembly, or the target structural layer is the proton exchange membrane, anode gas diffusion layer and cathode gas diffusion layer, anode catalyst layer and cathode catalyst layer of the membrane electrode assembly, etc.

[0054] The spatial domain thickness characteristic data is characteristic data used to characterize the nonlinear relationship between the plane position and the cross-sectional thickness of the target structure layer. For example, the spatial domain thickness characteristic data is a cross-sectional thickness curve of the target structure layer.

[0055] The frequency domain thickness characteristic data is characteristic data used to characterize the periodic relationship between the plane position distance and the cross-sectional thickness peak of the target structure layer. For example, the frequency domain thickness characteristic data is a cross-sectional thickness spectrum diagram of the target structure layer.

[0056] It should be noted that the terms "first", "second", "third", etc. mentioned in this application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. The term "and / or" mentioned in this application describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can be represented as A existing alone, A and B existing at the same time, and B existing alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0057] After introducing the technical terms involved in this application, the application scenarios and design concepts of this application are briefly introduced.

[0058] In a fuel cell system, the uniformity and attenuation of the thickness of the membrane electrode assembly at various locations are important indicators reflecting the degree of corrosion of the membrane electrode assembly. Detection of the uniformity and attenuation of the thickness of the membrane electrode assembly at various locations is crucial. The traditional solution is usually to measure the thickness of the membrane electrode assembly in full size through a microscope, and then analyze the uniformity and attenuation based on the measured full-size thickness of the membrane electrode assembly. This membrane electrode thickness detection solution has a long cycle, a large workload, and extremely high labor and equipment costs.

[0059] To this end, after obtaining the cross-sectional thicknesses at multiple different positions of the target structural layer of the membrane electrode assembly, the present application inputs the cross-sectional thicknesses at multiple different positions of the target structural layer into a membrane electrode thickness detection model, so as to determine the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structural layer based on the cross-sectional thicknesses at multiple different positions of the target structural layer through the membrane electrode thickness detection model, and then analyzes the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation of the target structural layer, thereby achieving automatic detection of the thickness uniformity and / or thickness attenuation by obtaining a limited number of cross-sectional thicknesses, thereby reducing the cycle and workload of membrane electrode thickness detection, and reducing labor and equipment detection costs, and by performing thickness detection based on the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structural layer through the membrane electrode thickness detection model, the accuracy of the thickness uniformity and / or thickness attenuation of the membrane electrode can be improved.

[0060] After introducing the application scenarios and design ideas of the present application, the technical solutions provided by the present application are described in detail below.

[0061] The present application provides a fuel cell membrane electrode thickness detection method, which can be applied to electronic devices such as computers, for example. Figure 2 As shown, the general process of the fuel cell membrane electrode thickness detection method provided in the embodiment of the present application is as follows:

[0062] Step 201: Obtain cross-sectional thicknesses at multiple different locations of a target structural layer of a membrane electrode assembly. In practical applications, after the membrane electrode assembly is subjected to working condition tests such as carbon corrosion, mechanical durability, and anti-reverse polarity of a small piece fixture, the thickness of each part of the membrane electrode assembly may change compared to the original sample. Therefore, for the membrane electrode assembly after the working condition test, the structure layer to be tested of the membrane electrode assembly can be configured as the target structure layer according to actual needs, for example, the membrane electrode assembly is configured as the target structure layer, and the proton exchange membrane of the membrane electrode assembly is configured as the target structure layer, and the proton exchange membrane, anode gas diffusion layer, and cathode gas diffusion layer of the membrane electrode assembly are configured as the target structure layer, and the proton exchange membrane, anode gas diffusion layer, cathode gas diffusion layer, anode catalyst layer, and cathode catalyst layer of the membrane electrode assembly are configured as the target structure layer, etc., and then, the cross-sectional thickness of multiple different positions calibrated according to the set size interval on the target structure layer of the membrane electrode assembly can be obtained. Specifically, when obtaining the cross-sectional thickness of multiple different positions calibrated according to the set size interval on the target structure layer of the membrane electrode assembly, but not limited to any of the following methods can be used:

[0063] The first method is to collect cross-sectional infrared sensing signals at multiple different positions on the target structure layer calibrated according to set size intervals through an infrared sensor, and determine the cross-sectional thickness at multiple different positions of the target structure layer based on the cross-sectional infrared sensing signals at multiple different positions of the target structure layer.

[0064] For example, multiple positions are pre-calibrated on the target structure layer of the membrane electrode assembly at a set size interval of 3 cm, and then the cross-sectional infrared sensing signals of the multiple positions are respectively collected by an infrared sensor, so as to determine the cross-sectional thicknesses of multiple different positions of the target structure layer based on the cross-sectional infrared sensing signals of the multiple positions.

[0065] The second method is to collect the cross-sectional depth image and cross-sectional RGB image of the target structure layer through a binocular camera device, and perform image analysis on the cross-sectional depth image and the cross-sectional RGB image to identify the cross-sectional thickness at multiple different positions on the target structure layer calibrated according to set size intervals.

[0066] For example, a cross-sectional depth image and a cross-sectional RGB image of a target structure layer are collected by a binocular camera device, and a target recognition model is used to identify the target structure layer in the cross-sectional depth image and the cross-sectional RGB image, as well as the cross-sectional thicknesses at multiple positions on the target structure layer that are pre-calibrated at a set size interval of 3 cm, thereby obtaining the cross-sectional thicknesses at multiple different positions of the target structure layer.

[0067] The third method is to measure the cross-sectional thickness at multiple different locations on the target structure layer calibrated at set size intervals through a microscope.

[0068] For example, multiple positions are pre-calibrated on the target structure layer of the membrane electrode assembly at a set size interval of 3 cm, and then the cross-sectional thickness of the multiple positions is measured by any microscope such as a metallographic microscope, a laser microscope or an electron microscope, thereby obtaining the cross-sectional thickness of multiple different positions of the target structure layer.

[0069] Step 202: input the cross-sectional thicknesses at multiple different positions of the target structure layer into the membrane electrode thickness detection model to obtain the thickness detection result of the target structure layer output by the membrane electrode thickness detection model; wherein the membrane electrode thickness detection model is used to determine the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, and then analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain the thickness detection result used to characterize the thickness uniformity and / or thickness attenuation degree of the target structure layer.

[0070] In practical applications, the membrane electrode thickness detection model includes multiple convolutional layers for obtaining spatial domain thickness feature data and frequency domain thickness feature data, and a fully connected layer connected to the multiple convolutional layers for obtaining thickness detection results. For example, see Figure 3 As shown, in one embodiment, the membrane electrode thickness detection model may include a first convolution layer for obtaining spatial domain thickness feature data and a second convolution layer for obtaining frequency domain thickness feature data, which are connected in sequence, and a fully connected layer for obtaining thickness detection results, which is respectively connected to the first convolution layer and the second convolution layer. Based on this, after the cross-sectional thicknesses at multiple different positions of the target structure layer are input into the membrane electrode thickness detection model, when the membrane electrode thickness detection model is used to obtain the thickness uniformity and / or thickness attenuation of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, the following methods may be used but are not limited to:

[0071] First, through the first convolution layer in the membrane electrode thickness detection model, the cross-sectional thickness at multiple different positions of the target structure layer is curve fitted to obtain the spatial domain thickness feature data used to characterize the nonlinear relationship between the plane position and the cross-sectional thickness of the target structure layer; wherein the spatial domain thickness feature data can be as follows Figure 4 The cross-sectional thickness curve of the target structure layer is shown.

[0072] Then, through the second convolution layer in the membrane electrode thickness detection model, the spatial domain thickness feature data is Fourier transformed to obtain the frequency domain thickness feature data used to characterize the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak value; wherein the frequency domain thickness feature data can be as follows Figure 5 The cross-sectional thickness spectrum of the target structure layer is shown.

[0073] Finally, through the fully connected layer in the membrane electrode thickness detection model, the spatial domain thickness characteristic data and the frequency domain thickness characteristic data are analyzed to obtain the periodic change data of the cross-sectional thickness of the target structure layer. Based on the periodic change data of the cross-sectional thickness of the target structure layer, the thickness detection result used to characterize the thickness uniformity and / or thickness attenuation of the target structure layer is determined.

[0074] In another embodiment, the membrane electrode thickness detection model may further include a first convolution layer for obtaining spatial domain thickness feature data, a third convolution layer for smoothing the spatial domain thickness feature data, and a second convolution layer for obtaining frequency domain thickness feature data, which are sequentially connected, and a fully connected layer for obtaining thickness detection results respectively connected to the first convolution layer and the second convolution layer. Based on this, after the cross-sectional thicknesses at multiple different positions of the target structure layer are input into the membrane electrode thickness detection model, when the membrane electrode thickness detection model is used to obtain the thickness uniformity and / or thickness attenuation of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, the following methods may be used but are not limited to:

[0075] First, through the first convolution layer in the membrane electrode thickness detection model, the cross-sectional thickness at multiple different positions of the target structure layer is curve fitted to obtain the spatial domain thickness feature data used to characterize the nonlinear relationship between the plane position and the cross-sectional thickness of the target structure layer; wherein the spatial domain thickness feature data can be as follows Figure 4 The cross-sectional thickness curve of the target structure layer is shown.

[0076] Then, the spatial domain thickness feature data of the target structure layer is smoothed by the third convolution layer in the membrane electrode thickness detection model to obtain smoothed spatial domain thickness feature data.

[0077] Secondly, through the second convolution layer in the membrane electrode thickness detection model, the smoothed spatial domain thickness feature data is Fourier transformed to obtain the frequency domain thickness feature data used to characterize the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak value; wherein the frequency domain thickness feature data can be as follows Figure 5 The cross-sectional thickness spectrum of the target structure layer is shown.

[0078] Finally, through the fully connected layer in the membrane electrode thickness detection model, the spatial domain thickness characteristic data and the frequency domain thickness characteristic data are analyzed to obtain the periodic change data of the cross-sectional thickness of the target structure layer. Based on the periodic change data of the cross-sectional thickness of the target structure layer, the thickness detection result used to characterize the thickness uniformity and / or thickness attenuation of the target structure layer is determined.

[0079] It is worth mentioning that in the embodiment of the present application, the spatial domain thickness characteristic data and the frequency domain thickness characteristic data are analyzed through the fully connected layer in the membrane electrode thickness detection model, and the periodic change data such as the peak values ​​of each cross-sectional thickness when the cross-sectional thickness of the target structure layer changes with the size position and the plane position distance between each two cross-sectional thickness peaks can be obtained. When detecting the thickness uniformity, the thickness uniformity used to characterize the target structure layer can be directly determined based on the thickness change period represented by the periodic change data. The longer the thickness change period represented by the periodic change data, the better the thickness uniformity of the target structure layer. The thickness uniformity can be a numerical value such as a score, a grade, or a percentage. When detecting the thickness attenuation degree, in addition to analyzing the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer after the operating condition test of the membrane electrode assembly through the fully connected layer in the membrane electrode thickness detection model to obtain the periodic change data of the cross-sectional thickness of the target structure layer after the operating condition test, it is also necessary to use the membrane electrode thickness detection model to detect the thickness attenuation degree. The fully connected layer in the measurement model analyzes the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structural layer of the membrane electrode assembly before the operating condition test to obtain the periodic change data of the cross-sectional thickness of the target structural layer before the operating condition test, thereby determining the thickness attenuation degree of the target structural layer based on the periodic change data of the cross-sectional thickness of the target structural layer after the membrane electrode assembly is tested under the operating condition and the periodic change data of the cross-sectional thickness of the target structural layer before the membrane electrode assembly is tested under the operating condition. Similarly, the thickness attenuation degree of the target structural layer can be a numerical value such as a score, a grade, or a percentage. Among them, the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structural layer of the membrane electrode assembly before the operating condition test are obtained by inputting the cross-sectional thickness of multiple different positions of the target structural layer of the membrane electrode assembly before the operating condition test into the membrane electrode thickness detection model, and then obtaining them through the first convolution layer (and the third convolution layer) and the second convolution layer in the membrane electrode thickness detection model. The acquisition method is the same as the above-described method, and the repeated parts will not be repeated.

[0080] The following is a brief description of the training process of the membrane electrode thickness detection model provided in the embodiment of the present application. Figure 6 As shown, the training process of the membrane electrode thickness detection model provided in the embodiment of the present application includes the following steps:

[0081] Step 601: Obtain a training sample set; wherein the training sample set includes a plurality of training sample data, each training sample data includes at least the cross-sectional thickness of a target structure layer of a membrane electrode assembly at a plurality of different positions before and after the operating condition test, as well as the standard thickness uniformity and standard thickness attenuation of the target structure layer.

[0082] Step 602: Train the initial membrane electrode thickness detection model based on the training sample set to obtain the membrane electrode thickness detection model. Specifically, the initial membrane electrode thickness detection model can be iteratively trained based on the training sample set until it is determined that the iteration termination condition is met, and the membrane electrode thickness detection model is obtained based on the model parameters of the initial membrane electrode thickness detection model updated when the model training operation is performed for the last time; wherein each execution of the model training operation includes: selecting target training sample data from the training sample set; inputting the cross-sectional thickness of the target structure layer of the membrane electrode assembly contained in the target training sample data at multiple different positions before and after the working condition test into the initial membrane electrode thickness detection model to obtain the predicted thickness uniformity and predicted thickness attenuation output by the initial membrane electrode thickness detection model; determining the current loss value based on the predicted thickness uniformity and predicted thickness attenuation and the standard thickness uniformity and standard thickness attenuation of the target structure layer contained in the target training sample data; and updating the model parameters of the initial membrane electrode thickness detection model based on the current loss value. In practical applications, the iteration termination condition may be, but is not limited to, the number of iterations being not less than the first threshold, the current loss value being not greater than the second threshold, etc.

[0083] Based on the above embodiments, the present application also provides a fuel cell membrane electrode thickness detection device, see Figure 7 As shown, the fuel cell membrane electrode thickness detection device 700 provided in the embodiment of the present application at least includes:

[0084] An acquisition unit 701 is used to acquire the cross-sectional thickness of a plurality of different positions of a target structural layer of a membrane electrode assembly;

[0085] The detection unit 702 is used to input the cross-sectional thicknesses at multiple different positions of the target structure layer into the membrane electrode thickness detection model to obtain the thickness detection result of the target structure layer output by the membrane electrode thickness detection model; wherein the membrane electrode thickness detection model is used to determine the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, and then analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain the thickness detection result used to characterize the thickness uniformity and / or thickness attenuation degree of the target structure layer.

[0086] In one possible implementation, the acquisition unit 701 is specifically used to collect cross-sectional infrared sensing signals at multiple different positions on the target structure layer calibrated according to set size intervals through an infrared sensor, and determine the cross-sectional thicknesses at multiple different positions of the target structure layer based on the cross-sectional infrared sensing signals at multiple different positions of the target structure layer; or; collect a cross-sectional depth image and a cross-sectional RGB image of the target structure layer through a binocular camera device, and perform image analysis on the cross-sectional depth image and the cross-sectional RGB image to identify the cross-sectional thicknesses at multiple different positions on the target structure layer calibrated according to set size intervals; or; measure the cross-sectional thicknesses at multiple different positions on the target structure layer calibrated according to set size intervals through a microscope.

[0087] In one possible implementation, the detection unit 702 is specifically used to perform curve fitting on the cross-sectional thicknesses at multiple different positions of the target structure layer to obtain spatial domain thickness characteristic data for characterizing the nonlinear relationship between the plane position and the cross-sectional thickness of the target structure layer; and perform Fourier transform on the spatial domain thickness characteristic data of the target structure layer to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak.

[0088] In a possible implementation, the detection unit 702 is specifically used to smooth the spatial domain thickness characteristic data of the target structure layer to obtain smoothed spatial domain thickness characteristic data; and to perform Fourier transform on the smoothed spatial domain thickness characteristic data to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance of the target structure layer and the cross-sectional thickness peak value.

[0089] In one possible implementation, the detection unit 702 is specifically used to analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain the periodic variation data of the cross-sectional thickness of the target structure layer; based on the periodic variation data of the cross-sectional thickness of the target structure layer, determine the thickness detection result used to characterize the thickness uniformity and / or thickness attenuation of the target structure layer.

[0090] In a possible implementation, the membrane electrode thickness detection model includes a plurality of convolutional layers for acquiring thickness feature data in the spatial domain and thickness feature data in the frequency domain, and a fully connected layer connected to the plurality of convolutional layers for acquiring thickness detection results.

[0091] In a possible implementation, the fuel cell membrane electrode thickness detection device 700 provided in the present application further includes:

[0092] The training unit 703 is used to obtain a training sample set, which includes a plurality of training sample data. Each training sample data includes the cross-sectional thickness of a target structure layer of a membrane electrode assembly at a plurality of different positions before and after the operating condition test, as well as the standard thickness uniformity and standard thickness attenuation of the target structure layer. The initial membrane electrode thickness detection model is trained based on the training sample set to obtain a membrane electrode thickness detection model.

[0093] It should be noted that the principle of solving the technical problem by the fuel cell membrane electrode thickness detection device 700 provided in the embodiment of the present application is similar to the fuel cell membrane electrode thickness detection method provided in the embodiment of the present application. Therefore, the implementation of the fuel cell membrane electrode thickness detection device 700 provided in the embodiment of the present application can refer to the implementation of the fuel cell membrane electrode thickness detection method provided in the embodiment of the present application, and the repeated parts will not be repeated.

[0094] After introducing the fuel cell membrane electrode thickness detection method and device provided in the embodiment of the present application, the electronic device provided in the embodiment of the present application is briefly introduced next.

[0095] The electronic device provided in the embodiment of the present application may be, but is not limited to, a computer, etc. For details, see Figure 8 As shown, the electronic device 800 provided in the embodiment of the present application includes at least a processor 801, a memory 802, and a computer program stored in the memory 802 and executable on the processor 801. When the processor 801 executes the computer program, the above-mentioned fuel cell membrane electrode thickness detection method provided in the embodiment of the present application is implemented.

[0096] The electronic device 800 provided in the embodiment of the present application may further include a bus 803 connecting different components (including the processor 801 and the memory 802). The bus 803 represents one or more of several types of bus structures, including a memory bus, a peripheral bus, a local bus, and the like.

[0097] The memory 802 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 8021 and / or a cache memory 8022, and may further include a read-only memory (ROM) 8023. The memory 802 may also include a program tool 8025 having a group (at least one) of program modules 8024, and the program modules 8024 include but are not limited to an operating subsystem, one or more application programs, other program modules, and program data, each of which or some combination may include the implementation of a network environment.

[0098] The processor 801 may be a processing element or a collective term for multiple processing elements. For example, the processor 801 may be a central processing unit (CPU) or one or more integrated circuits configured to implement the fuel cell membrane electrode thickness detection method provided in the embodiment of the present application. Specifically, the processor 801 may be a general-purpose processor, including but not limited to a CPU, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.

[0099] The electronic device 800 may also communicate with one or more external devices 804, such as one or more input devices (e.g., keyboards, remote controls, etc.), one or more devices that enable a user to interact with the electronic device 800 (e.g., mobile phones, computers, etc.), and devices that enable the electronic device 800 to communicate with one or more other electronic devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 805. In addition, the electronic device 800 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 806. Figure 8As shown, the network adapter 806 communicates with other modules of the electronic device 800 via the bus 803. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 800, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, disk arrays (Redundant Arrays of Independent Disks, RAID) subsystems, tape drives, and data backup storage subsystems.

[0100] It should be noted that Figure 8 The electronic device 800 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0101] In addition, the embodiment of the present application also provides a computer-readable storage medium, which stores computer instructions, and when the computer instructions are executed by the processor, the fuel cell membrane electrode thickness detection method provided in the embodiment of the present application is implemented. Specifically, the computer instructions can be built-in or installed in the processor, so that the processor can implement the above-mentioned fuel cell membrane electrode thickness detection method provided in the embodiment of the present application by executing the built-in or installed computer instructions.

[0102] In addition, the fuel cell membrane electrode thickness detection method provided in the embodiment of the present application can also be implemented as a program product, which includes a program code. When the program code is executed by a processor, the above-mentioned fuel cell membrane electrode thickness detection method provided in the embodiment of the present application is implemented.

[0103] The program product provided in the embodiments of the present application may adopt any combination of one or more readable media, wherein the readable medium may be a readable signal medium or a readable storage medium, and the readable storage medium may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any combination of the above. Specifically, more specific examples of readable storage media (a non-exhaustive list) include an electrical connection with one or more wires, a portable disk, a hard disk, RAM, ROM, Erasable Programmable Read Only Memory (EPROM), optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0104] The program product provided in the embodiment of the present application may adopt a CD-ROM and include program code, and may also be run on an electronic device. However, the program product provided in the embodiment of the present application is not limited thereto. In the embodiment of the present application, the readable storage medium may be any tangible medium containing or storing a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.

[0105] It should be noted that, although several units or subunits of the device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided into multiple units to be embodied.

[0106] In addition, although the operations of the method of the present application are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0107] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0108] Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if these modifications and variations of the embodiments of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A fuel cell membrane electrode thickness detection method, characterized in that: include: Obtaining cross-sectional thicknesses at multiple different locations of a target structural layer of a membrane electrode assembly; The cross-sectional thicknesses at multiple different positions of the target structure layer are input into a membrane electrode thickness detection model to obtain a thickness detection result of the target structure layer output by the membrane electrode thickness detection model; wherein the membrane electrode thickness detection model is used to determine the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, and then analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation of the target structure layer.

2. The fuel cell membrane electrode thickness detection method according to claim 1, characterized in that: Obtain the cross-sectional thickness of a target structural layer of a membrane electrode assembly at multiple different locations, including: Collecting cross-sectional infrared sensing signals of a plurality of different positions on the target structure layer calibrated at set size intervals by means of an infrared sensor, and determining the cross-sectional thickness of the plurality of different positions of the target structure layer based on the cross-sectional infrared sensing signals of the plurality of different positions of the target structure layer; or; Acquire a cross-sectional depth image and / or a cross-sectional RGB image of the target structure layer by a binocular camera device, and perform image analysis on the cross-sectional depth image and / or the cross-sectional RGB image to identify the cross-sectional thicknesses at a plurality of different positions on the target structure layer calibrated at set size intervals; or; The cross-sectional thickness of a plurality of different positions on the target structure layer calibrated at set size intervals is measured by a microscope.

3. The fuel cell membrane electrode thickness detection method according to claim 1, characterized in that: Determining the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer includes: Performing curve fitting on the cross-sectional thicknesses at a plurality of different positions of the target structure layer to obtain spatial domain thickness characteristic data for characterizing the nonlinear relationship between the plane position and the cross-sectional thickness of the target structure layer; The spatial domain thickness characteristic data of the target structure layer is subjected to Fourier transformation to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance and the cross-sectional thickness peak value of the target structure layer.

4. The fuel cell membrane electrode thickness detection method according to claim 3, characterized in that: Performing Fourier transform on the spatial domain thickness characteristic data of the target structure layer to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance and the cross-sectional thickness peak value of the target structure layer, including: Smoothing the spatial domain thickness characteristic data of the target structural layer to obtain smoothed spatial domain thickness characteristic data; The smoothed spatial domain thickness characteristic data is subjected to Fourier transformation to obtain frequency domain thickness characteristic data for characterizing the periodic relationship between the plane position distance and the cross-sectional thickness peak of the target structure layer.

5. The fuel cell membrane electrode thickness detection method according to claim 1, characterized in that: The spatial domain thickness characteristic data and the frequency domain thickness characteristic data are analyzed to obtain a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation of the target structure layer, including: Analyzing the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain periodic variation data of the cross-sectional thickness of the target structure layer; Based on the periodic variation data of the cross-sectional thickness of the target structural layer, a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation of the target structural layer is determined.

6. The fuel cell membrane electrode thickness detection method according to claim 1, characterized in that: The membrane electrode thickness detection model includes a plurality of convolutional layers for acquiring thickness feature data in the spatial domain and thickness feature data in the frequency domain, and a fully connected layer connected to the plurality of convolutional layers for acquiring thickness detection results.

7. The fuel cell membrane electrode thickness detection method according to any one of claims 1 to 6, characterized in that: Before obtaining the cross-sectional thickness of a plurality of different positions of the target structural layer of the membrane electrode assembly, the method further includes: Acquire a training sample set; wherein the training sample set includes a plurality of training sample data, each of which includes the cross-sectional thickness of a target structural layer of a membrane electrode assembly at a plurality of different positions before and after the operating condition test, and the standard thickness uniformity and standard thickness attenuation of the target structural layer; The membrane electrode thickness detection model is obtained by training an initial membrane electrode thickness detection model based on the training sample set.

8. A fuel cell membrane electrode thickness detection device, characterized in that: include: An acquisition unit, used for acquiring cross-sectional thicknesses at multiple different positions of a target structural layer of the membrane electrode assembly; A detection unit is used to input the cross-sectional thicknesses at multiple different positions of the target structure layer into a membrane electrode thickness detection model to obtain a thickness detection result of the target structure layer output by the membrane electrode thickness detection model; wherein the membrane electrode thickness detection model is used to determine the spatial domain thickness characteristic data and the frequency domain thickness characteristic data of the target structure layer based on the cross-sectional thicknesses at multiple different positions of the target structure layer, and then analyze the spatial domain thickness characteristic data and the frequency domain thickness characteristic data to obtain a thickness detection result for characterizing the thickness uniformity and / or thickness attenuation of the target structure layer.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the fuel cell membrane electrode thickness detection method as claimed in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the fuel cell membrane electrode thickness detection method according to any one of claims 1 to 7 is implemented.

Images