A fuel cell safety management method and device based on a magnetic field vector included angle

Abstract

This application discloses a fuel cell safety management method and device based on the magnetic field vector angle, relating to the field of fuel cell safety. The method includes: preprocessing the current magnetic induction intensity, and vector summing the preprocessed first, second, and third magnetic induction intensities to obtain the total magnetic induction intensity; calculating the angle between the total magnetic induction intensity and the preprocessed third magnetic induction intensity to obtain the target magnetic field vector angle; inputting the target magnetic field vector angle into a trained fuel cell safety management model to obtain the risk type and risk score of the target fuel cell; when the risk type of the target fuel cell is membrane dry risk or flooding risk, calculating the bypass valve opening amount based on the risk score corresponding to membrane dry risk or flooding risk, and controlling the throttling size of the bypass valve in the air supply system based on the bypass valve opening amount. This application achieves safe management of the fuel cell without damaging the original structure of the fuel cell.

Description

Technical Field

[0001] This application relates to the field of fuel cell safety technology, and in particular to a fuel cell safety management method and device based on the angle between magnetic field vectors. Background Technology

[0002] A fuel cell is a device that directly converts chemical energy into electrical energy. Its basic components include an anode, a cathode, and an electrolyte membrane. Fuel cells primarily generate electricity, water, and heat through the reaction of hydrogen and oxygen. Due to their high efficiency and low emissions, fuel cells are widely used in transportation, portable power sources, and stationary power generation.

[0003] With the rapid development of fuel cell technology, proton exchange membrane fuel cells (PEMFCs) are gradually gaining public attention. During fuel cell operation, the proton exchange membrane must be kept in a properly humidified state to achieve efficient proton conduction and effective isolation of reactant gases. However, in environments with high temperature, low humidity, or uneven fluid flow, the moisture in the proton exchange membrane may evaporate rapidly, leading to membrane drying. This dryness not only reduces the ionic conductivity of the proton exchange membrane but also increases the internal resistance of the cell. Conversely, in environments with high humidity or large fluid flows, excessive moisture in the proton exchange membrane may cause flooding, significantly reducing the fuel cell's output power and overall efficiency.

[0004] To address the aforementioned issues, effective water management of the proton exchange membrane (PEM) is typically used to ensure proton conductivity and overall battery efficiency. This effective water management is primarily achieved through the air supply system. The air supply system consists of an air filter, air compressor, intercooler, bypass valve (i.e., humidifier), and electronic throttle. The bypass valve's function is to humidify the air; properly controlling the bypass valve's throttling effectively regulates the dry / wet state of the PEM. To achieve safe management of the fuel cell and effectively control the opening of the bypass valve in the air supply system, current technologies typically employ invasive sensor placement methods to modify the fuel cell structure. Further monitoring of voltage and pressure data is used for fault diagnosis and safety control. However, this approach damages the original structure and performance of the fuel cell, hindering its safe and efficient operation. Summary of the Invention

[0005] The purpose of this application is to provide a method and device for the safety management of fuel cells based on the angle between magnetic field vectors, so as to achieve effective safety management of fuel cells without damaging the original structure of the fuel cells.

[0006] To achieve the above objectives, this application provides the following solution:

[0007] In a first aspect, this application provides a fuel cell safety management method based on the angle between magnetic field vectors, the fuel cell safety management method based on the angle between magnetic field vectors includes:

[0008] The magnetic flux density of the target fuel cell is obtained; the magnetic flux density includes: a first magnetic flux density, a second magnetic flux density, and a third magnetic flux density; the first magnetic flux density is the magnetic flux density generated by a first membrane current; the second magnetic flux density is the magnetic flux density generated by a second membrane current; the third magnetic flux density is the magnetic flux density generated by the main current; the first membrane current and the second membrane current are currents in two mutually perpendicular directions within the proton exchange membrane plane of the target fuel cell; the main current is the current from the anode to the cathode of the target fuel cell.

[0009] The magnetic flux density of the current is preprocessed, and the preprocessed first magnetic flux density, the preprocessed second magnetic flux density, and the preprocessed third magnetic flux density are vector summed to obtain the total magnetic flux density.

[0010] Calculate the angle between the total magnetic flux density and the pre-processed third magnetic flux density to obtain the target magnetic field vector angle;

[0011] The angle of the target magnetic field vector is input into the trained fuel cell safety management model to obtain the risk type and risk score of the target fuel cell; the trained fuel cell safety management model is built based on the LSTM model; the risk types include: no risk, membrane dry risk, and flooding risk;

[0012] When the risk type of the target fuel cell is membrane dryness risk or flooding risk, the bypass valve opening amount is calculated based on the risk score corresponding to membrane dryness risk or flooding risk, and the throttling size of the bypass valve in the air supply system is controlled based on the bypass valve opening amount.

[0013] In a second aspect, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the fuel cell safety management method based on magnetic field vector angle described in the first aspect.

[0014] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0015] This application does not disrupt the original structure of the fuel cell. Instead, it introduces a novel evaluation metric: the angle between the target magnetic field vectors. By calculating this metric and inputting it into a trained fuel cell safety management model, the risk type and risk score of the fuel cell can be determined. The risk type output by the model can identify whether the fuel cell faces membrane dryness or flooding risks. When these risks occur, the bypass valve opening can be calculated based on the risk score corresponding to different risk types, thereby controlling the throttling magnitude of the bypass valve. Based on this design, this application effectively protects the original structure and performance of the fuel cell, enabling not only risk diagnosis but also effective safety management based on the risk type. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a fuel cell safety management method based on the included magnetic field vector angle provided in this application embodiment;

[0018] Figure 2 This is a diagram illustrating the arrangement of the magnetic induction intensity sensor on the fuel cell according to an embodiment of this application.

[0019] Figure 3 A diagram illustrating the effect of the first membrane current and its magnetic field on an embodiment of this application;

[0020] Figure 4 The second membrane current and its magnetic field effect diagram provided for the embodiments of this application;

[0021] Figure 5 The diagram shows the effect of the third membrane current and its magnetic field in the embodiments of this application.

[0022] Figure 6 A geometrical diagram showing the relationship between the total magnetic flux density and individual magnetic flux densities provided for embodiments of this application;

[0023] Figure 7 A schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0024] Symbol explanation:

[0025] Anode-1, cathode-2, proton exchange membrane-3. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] The purpose of this application is to provide a method and device for the safety management of fuel cells based on the angle between magnetic field vectors, so as to achieve effective safety management of fuel cells without damaging the original structure of the fuel cells.

[0028] The safety management of fuel cells is inseparable from their water management system. Inside the porous electrodes of a fuel cell, water exists in three forms: water vapor, liquid water, and membrane water, and these different forms of water can transform into each other.

[0029] The presence of water vapor is crucial for maintaining humidity inside the fuel cell. It helps maintain proper wettability of the proton exchange membrane, promoting efficient ion conduction. However, excessive water vapor increases internal humidity, potentially leading to the accumulation of liquid water. Liquid water is primarily generated from the condensation of water vapor (originating from humidification via the bypass valve and desorption of water from the membrane). A suitable amount of liquid water helps maintain a humid environment inside the fuel cell, but excessive liquid water can clog gas channels and pores, hindering gas transport and thus reducing fuel cell performance. Membrane water refers to water bound to sulfonic acid group molecules, which is essential for ion conductivity.

[0030] To achieve water management in fuel cells, a bypass valve (i.e., a humidifier) ​​in the air supply system is typically used to control the humidity of the input gas. By controlling the humidity through the bypass valve, sufficient moisture in the system can be ensured to maintain the normal operation of the proton exchange membrane.

[0031] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] Example 1

[0033] This embodiment provides a fuel cell safety management method based on the angle between magnetic field vectors, such as... Figure 1 As shown, the fuel cell safety management method based on the angle between magnetic field vectors includes:

[0034] Step S1: Obtain the current magnetic induction intensity of the target fuel cell.

[0035] In this embodiment, step S1 specifically includes:

[0036] Step S11: As Figure 2 As shown, a magnetic induction intensity sensor A is installed on the left or right side of the target fuel cell to collect the first magnetic induction intensity. The first magnetic induction intensity is the magnetic induction intensity generated by the first membrane current. In this embodiment, the direction of the first membrane current is from the outside to the inside, and the direction of the magnetic field it generates can be found in [reference needed]. Figure 3 .

[0037] Step S12: As Figure 2 As shown, a magnetic induction intensity sensor B is deployed on the front or rear surface of the target fuel cell to collect the second magnetic induction intensity. The second magnetic induction intensity is the magnetic induction intensity generated by the second membrane current. In this embodiment, the direction of the second membrane current is from bottom to top, and the direction of the magnetic field it generates can be found in [reference needed]. Figure 4 .

[0038] Step S13: As Figure 2 As shown, a magnetic induction intensity sensor C is deployed on the upper or lower surface of the target fuel cell to collect the third magnetic induction intensity. The third magnetic induction intensity is the magnetic induction intensity generated by the main current, which is the current flowing from the anode 1 to the cathode 2 of the fuel cell. In this embodiment, its direction is from left to right, and the direction of the magnetic field it generates can be found in [reference needed]. Figure 5 .

[0039] In one preferred embodiment, the directions of the first membrane current and the second membrane current are perpendicular to each other within the proton exchange membrane (plane) 3 of the target fuel cell.

[0040] In a preferred embodiment, the magnetic induction intensity sensor is selected as NJK-5002C or HMC104X, and the magnetic induction intensity sensor can be fixed to the side wall of the target fuel cell by adhesive bonding.

[0041] In one preferred embodiment, the first magnetic induction intensity, the second magnetic induction intensity, and the third magnetic induction intensity of the target fuel cell are all vectors.

[0042] Step S2: Preprocess the current magnetic induction intensity, and then vector sum the preprocessed first magnetic induction intensity, the preprocessed second magnetic induction intensity, and the preprocessed third magnetic induction intensity to obtain the total magnetic induction intensity.

[0043] In this embodiment, the data collected by the multiple magnetic induction intensity sensors deployed on the target fuel cell cannot be directly used for subsequent analysis and processing. Instead, it needs to undergo filtering, feature extraction, and normalization to ensure the validity of the collected data and avoid interference from irrelevant signals. After the above preprocessing of each magnetic induction intensity data, the total magnetic induction intensity can be further calculated.

[0044] B = B v1+B v2 +B n ;

[0045] In the formula, B is the total magnetic flux density. v1 The first magnetic induction intensity after preprocessing, B v2 B represents the second magnetic induction intensity after preprocessing. n This represents the third magnetic flux density after preprocessing. Since both the collected and preprocessed magnetic flux densities are in vector form, the calculation of the total magnetic flux density is also a vector summation calculation.

[0046] Step S3: Calculate the angle between the total magnetic induction intensity and the preprocessed third magnetic induction intensity to obtain the target magnetic field vector angle.

[0047] In this embodiment, the geometric relationship between the total magnetic flux density and each individual magnetic flux density is shown in [reference]. Figure 6 The angle between the total magnetic flux density and the pre-processed third magnetic flux density is defined as the target magnetic field vector angle, and the formula for calculating this target magnetic field vector angle is as follows:

[0048] θ = arccos(B·B) n / |B|·|B n |);

[0049] In the formula, θ is the angle between the target magnetic field vectors, |B| is the magnitude of B, and |B n |For B n The model.

[0050] Theoretically, when the target magnetic field vector angle θ increases, the magnetic field generated by the membrane current (first membrane current and second membrane current) relatively increases, meaning the membrane current increases relative to the main current, the resistance on the proton exchange membrane decreases, and the water content of the fuel cell increases. Conversely, when the target magnetic field vector angle θ decreases, the membrane current decreases relative to the main current, the magnetic field generated by the membrane current relatively decreases, meaning the membrane current decreases relative to the main current, the resistance on the proton exchange membrane increases, and the water content of the fuel cell decreases. Furthermore, the change in the target magnetic field vector angle Δθ can also reflect the fault state of the fuel cell. When the fuel cell experiences a flooding fault, the change in the target magnetic field vector angle Δθ is greater than 0, and as the degree of flooding in the fuel cell intensifies, the change in the target magnetic field vector angle Δθ also gradually increases. When the fuel cell experiences a membrane dryness fault, the change in the target magnetic field vector angle Δθ is less than 0, and as the degree of membrane dryness in the fuel cell intensifies, the change in the target magnetic field vector angle Δθ also gradually increases.

[0051] Step S4: Input the included angle of the target magnetic field vector into the trained fuel cell safety management model to obtain the risk type and risk score of the target fuel cell.

[0052] In this embodiment, the risk types of the target fuel cell include: no risk, membrane dryness risk, and flooding risk. When the risk type is no risk, the risk score is less than 60; when the risk type is membrane dryness risk or flooding risk, the risk score is greater than or equal to 60.

[0053] Furthermore, the process of determining the trained fuel cell safety management model includes:

[0054] Step 101: By controlling the relative humidity of the gas input to the fuel cell, simulate different working environments of the fuel cell, and calibrate the risk type and risk score of the fuel cell under different working environments.

[0055] The relative humidity of the gas input to the fuel cell is controlled by setting up a humidifier, and the relative humidity can be directly read from the humidifier. Therefore, the risk score of the fuel cell needs to be calibrated according to the following formula:

[0056]

[0057] In the formula, Score is the risk score. K represents the relative humidity of the gas. s The fuel cell coefficient (related to the model of the fuel cell and the magnetic induction sensor; in this embodiment, K) is... s The reference value is 100).

[0058] Step 102: Obtain the current magnetic induction intensity of the fuel cell under different working environments, and preprocess the current magnetic induction intensity under different working environments.

[0059] The process of obtaining the current magnetic induction intensity of the fuel cell under different working environments and pre-processing it is exactly the same as steps S1 to S2 above, and will not be repeated here.

[0060] Step 103: Based on the current magnetic induction intensity under different working environments after preprocessing, calculate the target magnetic field vector angle under different working environments.

[0061] The process of calculating the angle between the target magnetic field vectors under different working environments is exactly the same as step S3 above, and will not be repeated here.

[0062] Step 104: Determine the fuel cell dataset and divide it into training and test sets proportionally.

[0063] The fuel cell dataset includes the target magnetic field vector angle, risk type, and risk score of fuel cells under different working environments. After the fuel cell dataset is constructed, 75% of the data in the fuel cell dataset will be used as the training set, and 25% of the data in the fuel cell dataset will be used as the test set.

[0064] Step 105: Construct the LSTM model.

[0065] In order to effectively capture the relationship between the angle of the target magnetic field vector and the risk type and risk score, the LSTM model was selected as the base model for subsequent training.

[0066] Step 106: Using the target magnetic field vector angle as input and the risk type and risk score as output, train and test the LSTM model using the training set and test set respectively.

[0067] Specifically, the data on the angle between the target magnetic field vectors are grouped into a 1×10×N vector (N is the number of groups) and used as the input to the LSTM model. The corresponding risk type and risk score are grouped into a 1×2 vector and used as the output of the LSTM model.

[0068] Step 107: When the LSTM model's accuracy in determining the risk type and risk score is greater than 95%, the LSTM model training ends, and the trained fuel cell safety management model is obtained.

[0069] The process involves first inputting the training set into the LSTM model for training, and then using the test set to correct the accuracy of the LSTM model in judging faults. The training of the model is complete when the accuracy of the LSTM model in judging the membrane dry risk or flood risk of fuel cells reaches more than 95%.

[0070] Step S5: When the risk type of the target fuel cell is membrane dry risk or flooding risk, calculate the bypass valve opening amount based on the risk score corresponding to membrane dry risk or flooding risk, and control the throttling size of the bypass valve in the air supply system based on the bypass valve opening amount.

[0071] The formula for calculating the bypass valve opening is as follows:

[0072] ε = K × Score;

[0073] In the formula, ε is the bypass valve opening amount, and K is the opening coefficient (related to the fuel cell model). The bypass valve opening amount determines the opening size of the bypass valve in the air supply system. As the risk score of the target fuel cell changes, the corresponding bypass valve opening amount also changes continuously, thereby directly controlling the throttling size of the bypass valve.

[0074] In one preferred embodiment, after the trained fuel cell safety management model outputs the risk type and risk score of the target fuel cell, the warning unit interface associated with the model output will display the risk type, risk score, and historical changes in the target magnetic field vector angle of the target fuel cell.

[0075] Based on the above analysis, the fuel cell safety management method based on the magnetic field vector angle does not require simulation software. Instead, it uses a large amount of effective data to train an LSTM model, making the model highly applicable. This method can achieve non-invasive fuel cell fault diagnosis without changing the internal structure of the fuel cell, and it has higher accuracy in diagnosing early-stage fuel cells.

[0076] Example 2

[0077] This embodiment provides a computer device, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 7 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores the current magnetic flux density data of the target fuel cell. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a fuel cell safety management method based on the angle of the magnetic field vector.

[0078] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0079] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0080] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0081] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0082] All actions involving the acquisition of signals, information, or data in this application are carried out in accordance with the relevant data protection laws and policies of the country where the application is located, and with the authorization of the owner of the relevant device.

[0083] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0084] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A fuel cell safety management method based on the angle between magnetic field vectors, characterized in that, The fuel cell safety management method based on the magnetic field vector angle includes: The magnetic flux density of the target fuel cell is obtained; the magnetic flux density includes: a first magnetic flux density, a second magnetic flux density, and a third magnetic flux density; the first magnetic flux density is the magnetic flux density generated by a first membrane current; the second magnetic flux density is the magnetic flux density generated by a second membrane current; the third magnetic flux density is the magnetic flux density generated by the main current; the first membrane current and the second membrane current are currents in two mutually perpendicular directions within the proton exchange membrane plane of the target fuel cell; the main current is the current from the anode to the cathode of the target fuel cell. The magnetic flux density of the current is preprocessed, and the preprocessed first magnetic flux density, the preprocessed second magnetic flux density, and the preprocessed third magnetic flux density are vector summed to obtain the total magnetic flux density. Calculate the angle between the total magnetic flux density and the pre-processed third magnetic flux density to obtain the target magnetic field vector angle; The angle of the target magnetic field vector is input into the trained fuel cell safety management model to obtain the risk type and risk score of the target fuel cell; the trained fuel cell safety management model is built based on the LSTM model; the risk types include: no risk, membrane dry risk, and flooding risk; When the risk type of the target fuel cell is membrane dryness risk or flooding risk, the bypass valve opening amount is calculated based on the risk score corresponding to membrane dryness risk or flooding risk, and the throttling size of the bypass valve in the air supply system is controlled based on the bypass valve opening amount.

2. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, Obtaining the current magnetic flux density of the target fuel cell specifically includes: A magnetic induction intensity sensor is deployed on the left or right side of the target fuel cell to collect the first magnetic induction intensity. A magnetic induction intensity sensor is deployed on the front or rear surface of the target fuel cell to collect the second magnetic induction intensity. A magnetic induction intensity sensor is deployed on the upper or lower surface of the target fuel cell to collect the third magnetic induction intensity.

3. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, The first, second, and third magnetic induction intensities of the target fuel cell are all vectors.

4. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, The preprocessing of the magnetic flux density of the current includes at least filtering and normalization.

5. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, The formula for calculating the total magnetic flux density is: B=B v1 +B v2 +B n ; In the formula, B is the total magnetic flux density. v1 The first magnetic induction intensity after preprocessing, B v2 B represents the second magnetic induction intensity after preprocessing. n The third magnetic induction intensity is the pre-processed value.

6. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, The formula for calculating the angle between the target magnetic field vectors is: θ=arccos(B·B n / |B|·|B n |); In the formula, θ is the angle between the target magnetic field vectors, and B is the total magnetic induction intensity. n is the third magnetic induction intensity after preprocessing, and |·| is the modulus.

7. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, When the risk type is no risk, the risk score is less than 60; when the risk type is membrane dryness risk or flooding risk, the risk score is greater than or equal to 60.

8. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, The process of determining the trained fuel cell safety management model includes: By controlling the relative humidity of the gas input to the fuel cell, different working environments of the fuel cell are simulated, and the risk types and risk scores of the fuel cell under different working environments are calibrated. The magnetic flux density of the fuel cell under different operating conditions is obtained, and the magnetic flux density of the fuel cell under different operating conditions is preprocessed. Based on the pre-processed magnetic flux density under different working environments, the included angle of the target magnetic field vector under different working environments is calculated. A fuel cell dataset is determined and divided into a training set and a test set according to a certain ratio; the fuel cell dataset includes the target magnetic field vector angle, risk type, and risk score of the fuel cell under different working environments. Construct an LSTM model; The LSTM model is trained and tested using the target magnetic field vector angle as input and the risk type and risk score as output, respectively, using the training set and the test set. When the LSTM model achieves an accuracy of more than 95% in determining the risk type and risk score, the LSTM model training ends, and a trained fuel cell safety management model is obtained.

9. The fuel cell safety management method based on the magnetic field vector angle according to claim 1, characterized in that, The formula for calculating the bypass valve opening is: ε = K × Score; In the formula, ε is the bypass valve opening degree, K is the opening coefficient, and Score is the risk score.

10. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the fuel cell safety management method based on the magnetic field vector angle as described in any one of claims 1-9.

Images