Method and device for battery temperature field simulation by using machine learning and medium
By using a neural network model based on machine learning for battery temperature field simulation, the problems of long calculation time and high cost of traditional numerical methods are solved, and efficient simulation of battery temperature field is achieved.
Patent Information
- Application Number
- CN202511293486.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-11
- Publication Date
- 2025-12-23
AI Technical Summary
Traditional numerical methods for evaluating battery thermal performance suffer from long computation times and high costs.
The machine learning approach is used to obtain the battery temperature field simulation parameters and then use a pre-trained neural network model to simulate the battery temperature field. The termination conditions include simulation duration and temperature field change rate. The training data includes various battery geometric dimensions, convective heat transfer coefficient, ambient temperature, and heat generation rate.
This improves the efficiency of battery temperature field simulation and solves the problems of long calculation time and high cost in traditional methods.
Smart Images
Figure CN121189151A_ABST
Abstract
Description
Technical Field
[0001] This application relates to methods, apparatus, and media for simulating battery temperature fields using machine learning, and belongs to the field of computer technology. Background Technology
[0002] Battery temperature has a significant impact on battery performance, lifespan, and safety during use, and battery temperature simulation is an important part of the battery design process.
[0003] Traditional temperature simulation methods include numerical simulation. This involves establishing a mathematical model of the battery's heat transfer processes, such as conduction, convection, and radiation, based on the battery's physical characteristics and operating principles; determining the battery's boundary conditions (such as ambient temperature and heat dissipation conditions) and initial conditions (such as initial temperature distribution); solving the established thermal model using numerical methods to obtain the battery's temperature distribution under different operating conditions; and analyzing the simulation results to evaluate the battery's thermal performance, providing a basis for battery design and thermal management.
[0004] However, evaluating the thermal performance of batteries using numerical methods is time-consuming and costly. Summary of the Invention
[0005] This application provides a method, apparatus, and medium for simulating battery temperature fields using machine learning, which can solve the problems of long calculation time and high cost of traditional numerical methods for evaluating battery thermal performance. This application provides the following technical solution:
[0006] In a first aspect, a method for simulating battery temperature fields using machine learning is provided, the method comprising:
[0007] The temperature field simulation parameters of the battery are obtained, including the battery's geometric dimensions, the temperature field at the current moment, the convective heat transfer coefficient of each surface of the battery, the ambient temperature, the thermal conductivity, and the heat generation rate.
[0008] The temperature field simulation parameters are input into a pre-trained temperature field simulation model to obtain the temperature field corresponding to the next moment. The temperature field corresponding to the next moment is used as the new temperature field corresponding to the current moment. The above operation is repeated until the termination condition is met. The termination condition includes the simulation duration reaching a preset duration and / or the temperature field change rate being lower than a preset threshold. The temperature field simulation model is established based on a neural network and trained using training data. The training data includes sample data, which includes: various sample geometric dimensions of the battery, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate.
[0009] Optionally, the temperature field simulation model is obtained by using temperature field data solved by numerical methods as the true value for data-driven training; or, it is obtained by using the residual of the heat transfer equation as the loss function for physical-informed training; or, it is obtained by combining the temperature field as the true value and the loss function of the residual of the heat transfer equation for training.
[0010] Optionally, the training data further includes the temperature field data for the next time step corresponding to each set of sample data; correspondingly, the temperature field data obtained by numerical methods is used as the ground truth for data-driven training, including:
[0011] The temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate are input into the neural network to obtain the network output results.
[0012] The neural network is trained using the average absolute error between the network output and the temperature field at the next moment as the loss function.
[0013] Optionally, the step of using the residuals of the heat transfer equation as a loss function for physically informed training includes:
[0014] The temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate are input into the neural network to obtain the network output results.
[0015] The residuals of the heat transfer equation are determined based on the network output and the heat transfer equation, and the neural network is trained based on the residuals.
[0016] Optionally, the step of combining the temperature field data as the true value and the loss function of the residual of the heat transfer equation for training includes:
[0017] The temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate are input into the neural network to obtain the network output results.
[0018] Obtain the mean absolute error between the network output and the temperature field data at the next moment;
[0019] The residuals of the heat transfer equations are determined based on the network output results and the heat transfer equations.
[0020] The loss function is determined based on the weighted sum of the mean absolute error and the residuals to train the neural network.
[0021] Optionally, the heat production rate and the sample heat production rate are obtained by electrochemical calculation at each time step; or, a curve showing the change between the heat production rate and time is given in advance, and the value at each time step is extracted from the curve to obtain the heat production rate or the sample heat production rate.
[0022] Secondly, an apparatus for simulating battery temperature fields using machine learning is provided, the apparatus comprising:
[0023] The parameter acquisition module is used to acquire the temperature field simulation parameters of the battery. The temperature field simulation parameters include the geometric dimensions of the battery, the temperature field at the current moment, the convective heat transfer coefficient of each surface of the battery, the ambient temperature, the thermal conductivity of the battery, and the heat generation rate.
[0024] The temperature field simulation module is used to input the temperature field simulation parameters into a pre-trained temperature field simulation model to obtain the temperature field corresponding to the next moment; the temperature field corresponding to the next moment is used as the new temperature field corresponding to the current moment, and the above operation is repeated until the termination condition is met. The termination condition includes the simulation duration reaching a preset duration and / or the temperature field change rate being lower than a preset threshold. The temperature field simulation model is established based on a neural network and trained using training data. The training data includes sample data, which includes: various sample geometric dimensions of the battery, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate.
[0025] Optionally, the temperature field simulation model is obtained by using temperature field data solved by numerical methods as the true value for data-driven training; or, it is obtained by using the residual of the heat transfer equation as the loss function for physical-informed training; or, it is obtained by combining temperature field data as the true value and the loss function of the residual of the heat transfer equation for training.
[0026] Thirdly, an apparatus for simulating battery temperature field using machine learning is provided, the apparatus comprising a processor and a memory; the memory stores a program which is loaded and executed by the processor to implement the method for simulating battery temperature field using machine learning as described in the first aspect.
[0027] Fourthly, a computer-readable storage medium is provided, wherein a program is stored therein, the program being loaded and executed by the processor to implement the method for simulating battery temperature field using machine learning as described in the first aspect.
[0028] The beneficial effects of this application are as follows: By obtaining the temperature field simulation parameters of the battery, including the battery's geometric dimensions, the temperature field at the current moment, the convective heat transfer coefficients of various battery surfaces, the ambient temperature, the thermal conductivity, and the heat generation rate; inputting the temperature field simulation parameters into a pre-trained temperature field simulation model to obtain the temperature field at the next moment; using the temperature field at the next moment as the new temperature field at the current moment, repeating the above operation until the termination condition is met; the temperature field simulation model is established based on a neural network and trained using training data, including sample data, which includes: various sample geometric dimensions of the battery, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate; temperature field simulation under arbitrary operating conditions can be performed through neural networks, solving the problems of long calculation time and high cost of traditional numerical methods for evaluating the thermal performance of batteries; since the temperature field simulation model is pre-trained, the simulation efficiency of the battery temperature field can be improved.
[0029] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings. Attached Figure Description
[0030] Figure 1 This is a flowchart of a method for simulating battery temperature field using machine learning, provided in one embodiment of this application;
[0031] Figure 2 This is a block diagram of an apparatus for simulating battery temperature field using machine learning, provided in one embodiment of this application.
[0032] Figure 3 This is a block diagram of an apparatus for simulating battery temperature field using machine learning, provided in one embodiment of this application. Detailed Implementation
[0033] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.
[0034] Figure 1 This is a flowchart illustrating a method for simulating battery temperature field using machine learning, provided in one embodiment of this application. This embodiment uses the application of this method in an electronic device as an example for explanation. The electronic device can be a user terminal or a server. The user terminal can be a computer, tablet computer, mobile phone, etc. This embodiment does not limit the implementation method of the electronic device. The method includes at least the following steps:
[0035] Step 101: Obtain the temperature field simulation parameters of the battery. These temperature field simulation parameters include the battery's geometric dimensions, the temperature field at the current moment, the convective heat transfer coefficient of each surface of the battery, the ambient temperature, the thermal conductivity, and the heat generation rate.
[0036] The battery's geometric dimensions include its length, height, and thickness. Each battery has fixed geometric dimensions, thermal conductivity, and convective heat transfer coefficient. These parameters can be obtained through relevant battery parameters or received via a human-machine interface. This embodiment does not limit the method of obtaining the battery's geometric dimensions, thermal conductivity, and convective heat transfer coefficient.
[0037] The heat generation rate and the sample heat generation rate mentioned below vary with the battery's operating time. Based on this, the heat generation rate and the sample heat generation rate are obtained by solving the electrochemical method at each time step; or, a curve of the change between the heat generation rate and time is given in advance, and the value at each time step is extracted from the curve to obtain the heat generation rate or the sample heat generation rate.
[0038] Step 102: Input the temperature field simulation parameters into the pre-trained temperature field simulation model to obtain the temperature field corresponding to the next moment; use the temperature field corresponding to the next moment as the new temperature field corresponding to the current moment, and repeat the above operation until the termination condition is met. The termination condition includes the simulation duration reaching a preset duration and / or the temperature field change rate being lower than a preset threshold. The temperature field simulation model is established based on a neural network and trained using training data. The training data includes sample data, which includes: various sample geometric dimensions of the battery, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate.
[0039] The neural networks include, but are not limited to, network structures such as: Backpropagation Neural Network (BP) model, Transformer model, Deep Operator Network (deepONet), and Fourier Neural Operator (FNO). This embodiment does not limit the implementation method of the neural network.
[0040] In this embodiment, the temperature field simulation model is obtained by using the temperature field data solved by numerical methods as the true value for data-driven training; or, it is obtained by using the residual of the heat transfer equation as the loss function for physical-informed training; or, it is obtained by using the temperature field data as the true value and the loss function of the residual of the heat transfer equation for combined training.
[0041] In one example, the training data also includes the temperature field data for the next time step corresponding to each set of sample data; accordingly, the temperature field data obtained by numerical methods is used as the true value for data-driven training, including: inputting the temperature field at a certain time step, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate into the neural network to obtain the network output; and training the neural network based on the mean absolute error between the network output and the temperature field at the next time step as the loss function.
[0042] Taking a square battery as an example, the battery is 173mm long, 204mm high, and 72mm thick, with a convective heat transfer coefficient of 25W / (m²) for each surface. 2 The ambient temperature was 25℃, and the battery's thermal conductivity was 18 W / (m·K). Electrochemical-thermal coupled simulations were performed using traditional methods to generate temperature field simulation data at various charge / discharge rates for different battery geometries, convective heat transfer coefficients, ambient temperatures, and battery thermal conductivityes. 60% of the time points in the dataset were randomly selected as the training set, 30% as the validation set, and 10% as the test set.
[0043] Assuming we choose an FNO neural network, we take the temperature field at a certain moment, the battery's geometry, the convective heat transfer coefficient, the ambient temperature, the thermal conductivity, and the heat generation rate as inputs to the neural network, and use the average absolute error between the neural network's output and the temperature field at the next moment in the dataset as the loss function to train the neural network.
[0044] The trained neural network, when used, takes the initial temperature field, battery geometry, convective heat transfer coefficient, ambient temperature, thermal conductivity, and the initial heat generation rate obtained from solving the electrochemical equation as input to obtain the temperature field at the next moment. This temperature field is then used to solve the electrochemical equation again to obtain the heat generation rate at the next moment. The first temperature field at that moment, the battery geometry, convective heat transfer coefficient, ambient temperature, thermal conductivity, and the heat generation rate obtained from solving the electrochemical equation are then used as input to the neural network to obtain the second temperature field at the next moment after that. This process is repeated until a specified stopping condition for the battery is met, for example, a time limit of 3600 seconds.
[0045] In another example, the residuals of the heat transfer equation are used as a loss function for physically informed training, including: inputting the temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate into the neural network to obtain the network output; determining the residuals of the heat transfer equation based on the network output and the heat transfer equation, and training the neural network based on the residuals.
[0046] In another example, temperature field data is used as a loss function to combine the ground truth and the residuals of the heat transfer equation for training. This includes: inputting the temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate into the neural network to obtain the network output; obtaining the mean absolute error between the network output and the temperature field at the next moment; determining the residuals of the heat transfer equation based on the network output and the heat transfer equation; and determining the loss function based on the weighted sum of the mean absolute error and the residuals to train the neural network.
[0047] Taking a square battery as an example, the battery is 173mm long, 204mm high, and 72mm thick, with a convective heat transfer coefficient of 25W / (m²) for each surface. 2 The ambient temperature was 25℃, and the battery's thermal conductivity was 18 W / (m·K). Electrochemical-thermal coupled simulations were performed using traditional methods to generate temperature field simulation data at various charge / discharge rates for different battery geometries, convective heat transfer coefficients, ambient temperatures, and battery thermal conductivityes. 60% of the time points in the dataset were randomly selected as the training set, 30% as the validation set, and 10% as the test set.
[0048] Assuming we choose an FNO neural network, we take the temperature field at a certain moment, the battery's geometry, the convective heat transfer coefficient, the ambient temperature, the thermal conductivity, and the heat generation rate as inputs to the neural network. We use the sum of the average absolute error between the output of the neural network and the temperature field at the next moment in the dataset, and the residuals of the heat transfer equation as the loss function to train the neural network.
[0049] The trained neural network, when used, takes the initial temperature field, battery geometry, convective heat transfer coefficient, ambient temperature, thermal conductivity, and the initial heat generation rate obtained from solving the electrochemical equation as input to obtain the temperature field at the next time step. This temperature field is then used to solve the electrochemical equation again to obtain the heat generation rate at the next time step. This process is repeated until a specified stopping condition for the battery is met, such as a time limit of 3600 seconds.
[0050] In summary, the method for simulating battery temperature field using machine learning provided in this embodiment obtains battery temperature field simulation parameters, including the battery's geometric dimensions, the temperature field at the current moment, the convective heat transfer coefficients of various battery surfaces, ambient temperature, thermal conductivity, and heat generation rate. These parameters are then input into a pre-trained temperature field simulation model to obtain the temperature field at the next moment. The temperature field simulation model is based on a neural network and trained using training data, including sample data such as various sample battery geometric dimensions, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate. This method can solve the problems of long calculation time and high cost associated with traditional numerical methods for evaluating battery thermal performance. Furthermore, because the temperature field simulation model is pre-trained, the simulation efficiency of the battery temperature field can be improved.
[0051] Figure 2 This is a block diagram of an apparatus for simulating battery temperature field using machine learning, according to one embodiment of this application. The apparatus includes at least the following modules: a parameter acquisition module 210 and a temperature field simulation module 220.
[0052] The parameter acquisition module 210 is used to acquire the temperature field simulation parameters of the battery. The temperature field simulation parameters include the geometric dimensions of the battery, the temperature field at the current moment, the convective heat transfer coefficient of each surface of the battery, the ambient temperature, the thermal conductivity of the battery, and the heat generation rate.
[0053] The temperature field simulation module 220 is used to input the temperature field simulation parameters into a pre-trained temperature field simulation model to obtain the temperature field corresponding to the next moment; the temperature field corresponding to the next moment is used as the new temperature field corresponding to the current moment, and the above operation is repeated until the termination condition is met. The termination condition includes the simulation duration reaching a preset duration and / or the temperature field change rate being lower than a preset threshold. The temperature field simulation model is established based on a neural network and trained using training data. The training data includes sample data, which includes: various sample geometric dimensions of the battery, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate.
[0054] Optionally, the temperature field simulation model is obtained by using temperature field data solved by numerical methods as the true value for data-driven training; or, it is obtained by using the residual of the heat transfer equation as the loss function for physical-informed training; or, it is obtained by combining temperature field data as the true value and the loss function of the residual of the heat transfer equation for training.
[0055] For relevant details, please refer to the above method implementation examples.
[0056] It should be noted that the device for simulating battery temperature fields using machine learning provided in the above embodiments is only illustrated by the division of the functional modules described above. In practical applications, the functions can be assigned to different functional modules as needed, that is, the internal structure of the device for simulating battery temperature fields using machine learning can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the device for simulating battery temperature fields using machine learning provided in the above embodiments and the method embodiments for simulating battery temperature fields using machine learning belong to the same concept, and their specific implementation process is detailed in the method embodiments, which will not be repeated here.
[0057] Figure 3 This is a block diagram of an apparatus for simulating battery temperature fields using machine learning, according to one embodiment of this application. The apparatus includes at least a processor 301 and a memory 302.
[0058] Processor 301 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 301 may be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), or PLA (Programmable Logic Array). Processor 301 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 301 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 301 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.
[0059] Memory 302 may include one or more computer-readable storage media, which may be non-transitory. Memory 302 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in memory 302 is used to store at least one instruction, which is executed by processor 301 to implement the method for simulating battery temperature fields using machine learning provided in the method embodiments of this application.
[0060] In some embodiments, the apparatus for simulating battery temperature fields using machine learning may also optionally include: a peripheral device interface and at least one peripheral device. The processor 301, memory 302, and peripheral device interface can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface via a bus, signal line, or circuit board. Indicatively, peripheral devices include, but are not limited to: radio frequency circuitry, a touch display screen, audio circuitry, and a power supply.
[0061] Of course, the apparatus for simulating battery temperature field using machine learning may include fewer or more components, and this embodiment does not limit this.
[0062] Optionally, this application also provides a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the method of using machine learning to simulate battery temperature field in the above-described method embodiments.
[0063] Optionally, this application also provides a computer product including a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the method of using machine learning to simulate battery temperature field in the above-described method embodiments.
[0064] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0065] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.
Claims
1. A method of battery temperature field simulation using machine learning, characterized by, The method includes: The temperature field simulation parameters of the battery are obtained, including the battery's geometric dimensions, the temperature field at the current moment, the convective heat transfer coefficient of each surface of the battery, the ambient temperature, the thermal conductivity, and the heat generation rate. The temperature field simulation parameters are input into a pre-trained temperature field simulation model to obtain the temperature field corresponding to the next moment. The temperature field corresponding to the next moment is used as the new temperature field corresponding to the current moment. The above operation is repeated until the termination condition is met. The termination condition includes the simulation duration reaching a preset duration and / or the temperature field change rate being lower than a preset threshold. The temperature field simulation model is based on a neural network and trained using training data. The training data includes sample data, which includes: various sample geometric dimensions of the battery, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate.
2. The method of claim 1, wherein, The temperature field simulation model is obtained by using temperature field data solved by numerical methods as the true value and performing data-driven training. Alternatively, the residuals of the heat transfer equation can be used as the loss function to obtain the result through physical awareness training. Alternatively, it can be obtained by combining temperature field data as the true value and the loss function of the residual of the heat transfer equation for training.
3. The method of claim 2, wherein, The training data also includes the temperature field data for the next time step corresponding to each set of sample data; correspondingly, the temperature field data obtained by numerical methods is used as the ground truth for data-driven training, including: The temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate are input into the neural network to obtain the network output results. The neural network is trained using the average absolute error between the network output and the temperature field at the next moment as the loss function.
4. The method of claim 2, wherein, The training using the residuals of the heat transfer equation as a loss function for physical awareness includes: The temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate are input into the neural network to obtain the network output results. The residuals of the heat transfer equation are determined based on the network output and the heat transfer equation, and the neural network is trained based on the residuals.
5. The method of claim 2, wherein, The training process, which combines temperature field data as the true value and the loss function of the residuals of the heat transfer equation, includes: The temperature field at a certain moment, the sample geometry of the battery, the sample convective heat transfer coefficient, the sample ambient temperature, the sample thermal conductivity, and the sample heat generation rate are input into the neural network to obtain the network output results. Obtain the mean absolute error between the network output and the temperature field at the next moment; The residuals of the heat transfer equations are determined based on the network output results and the heat transfer equations. The loss function is determined based on the weighted sum of the mean absolute error and the residuals to train the neural network.
6. The method according to any one of claims 1 to 5, characterized in that, The heat production rate and the sample heat production rate are obtained by electrochemical calculation at each time step; or, a curve showing the change between the heat production rate and time is given in advance, and the value at each time step is extracted from the curve to obtain the heat production rate or the sample heat production rate.
7. A device for simulating battery temperature field using machine learning, characterized in that, The device includes: The parameter acquisition module is used to acquire the temperature field simulation parameters of the battery. The temperature field simulation parameters include the geometric dimensions of the battery, the temperature field at the current moment, the convective heat transfer coefficient of each surface of the battery, the ambient temperature, the thermal conductivity of the battery, and the heat generation rate. The temperature field simulation module is used to input the temperature field simulation parameters into a pre-trained temperature field simulation model to obtain the temperature field corresponding to the next moment; the temperature field corresponding to the next moment is used as the new temperature field corresponding to the current moment, and the above operation is repeated until the termination condition is met. The termination condition includes the simulation duration reaching a preset duration and / or the temperature field change rate being lower than a preset threshold. The temperature field simulation model is established based on a neural network and trained using training data. The training data includes sample data, which includes: various sample geometric dimensions of the battery, various sample convective heat transfer coefficients, sample ambient temperature, sample thermal conductivity, and sample heat generation rate.
8. The apparatus according to claim 7, characterized in that, The temperature field simulation model is obtained by using temperature field data solved by numerical methods as the true value and performing data-driven training. Alternatively, the residuals of the heat transfer equation can be used as the loss function to obtain the result through physical awareness training. Alternatively, it can be obtained by combining temperature field data as the true value and the loss function of the residual of the heat transfer equation for training.
9. An apparatus for simulating battery temperature field using machine learning, characterized in that, The device includes a processor and a memory; the memory stores a program that is loaded and executed by the processor to implement the method for simulating battery temperature field using machine learning as described in any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that, The storage medium stores a program that, when executed by a processor, is used to implement the method for simulating battery temperature field using machine learning as described in any one of claims 1 to 6.