Neural network-based power battery thermal management simulation method and system

Abstract

The present application relates to the technical field of model prediction, and particularly relates to a power battery thermal management simulation method and system based on a neural network. The method comprises the following steps: obtaining corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of a power battery through coupling experiment measurement, and performing three-dimensional coupling simulation construction to generate a three-dimensional physical field coupling model of the power battery; performing battery thermal characteristic evaluation according to the three-dimensional physical field coupling model of the power battery from four scales of corresponding micro materials, battery monomers, battery modules and thermal management systems of the power battery to obtain a power battery thermal characteristic data set; performing thermal management prediction training and thermal management constraint optimization control based on a neural network combined with the power battery thermal characteristic data set to construct a multi-scale deep neural network architecture, and generating an optimal thermal management control strategy of the power battery to perform corresponding thermal management control work. The present application can realize efficient simulation and intelligent prediction optimization of power battery thermal management.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of model prediction, in particular to a power battery thermal management simulation method and system based on neural network. BACKGROUND

[0002] With the wide application of electric vehicles and renewable energy systems, the thermal management of power batteries as the core energy storage unit has become a key factor affecting the performance, life and safety of the batteries. During the charging and discharging process of power batteries, especially when operating at high power density and high temperature environment, a large amount of heat is easily generated. If the heat cannot be effectively dissipated or managed, it will lead to high battery temperature, and further cause performance degradation, shortened life and even safety accidents.

[0003] In recent years, artificial intelligence, especially deep learning, has gradually attracted attention in physical modeling and engineering optimization. Based on neural network simulation method, especially deep neural network (DNN) and convolutional neural network (CNN) model, the non-linear characteristics and potential laws in complex systems can be learned through a large amount of historical data training. These methods can provide accurate thermal management simulation results in a short time, significantly reduce the calculation cost, and have strong adaptive ability to real-time respond to the thermal management needs of the battery under different working conditions. However, the existing power battery thermal management simulation method relies on a simple physical model, while the thermal behavior of power batteries involves electrochemical reactions, heat conduction, fluid flow and other multi-physical field coupling. The existing model cannot accurately describe the complex interaction process, and it is difficult to adapt to the dynamic changes of battery temperature in real time, thereby reducing the effect of power battery thermal management. SUMMARY

[0004] Therefore, it is necessary to provide a power battery thermal management simulation method and system based on neural network to solve at least one of the above technical problems.

[0005] To achieve the above purpose, a power battery thermal management simulation method based on neural network comprises the following steps:

[0006] Step S1: Obtain the corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery through coupling experiment measurement, and construct a three-dimensional coupling simulation of the power battery based on the corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery, to generate a three-dimensional physical field coupling model of the power battery;

[0007] Step S2: Perform battery thermal characteristic evaluation on four scales of micro material, battery monomer, battery module and thermal management system corresponding to the power battery through the three-dimensional physical field coupling model of the power battery to obtain the power battery thermal characteristic data set, including micro material thermal physical property data, battery monomer thermal characteristic data, battery module thermal distribution data and battery system thermal management parameter data;

[0008] Step S3: Construct a multi-scale deep neural network architecture based on the neural network and the power battery thermal characteristic data set, and perform thermal management prediction training on the multi-scale deep neural network architecture based on the power battery thermal characteristic data set to generate a power battery thermal management prediction model and output a power battery corresponding thermal management prediction result, including battery temperature distribution, thermal generation efficiency and thermal transfer coefficient corresponding thermal management prediction parameters;

[0009] Step S4: Perform thermal management constraint optimization control on the thermal management system corresponding to the power battery based on the thermal management prediction result of the power battery to generate an optimal thermal management control strategy for the power battery; upload the optimal thermal management control strategy of the power battery to the cloud platform to respond to the thermal management system corresponding to the power battery to perform corresponding thermal management control work.

[0010] Further, step S1 includes the following steps:

[0011] Step S11: Obtain the electrical performance parameters, thermal performance parameters and mechanical performance parameters corresponding to the power battery through coupling experimental measurement;

[0012] Step S12: Obtain the design structure data corresponding to each component of the power battery, and construct the topology connection of the power battery based on the design structure data corresponding to each component of the power battery to generate a topology connection architecture of each component of the power battery;

[0013] Step S13: Perform physical field simulation analysis on the power battery according to the electrical performance parameters, thermal performance parameters and mechanical performance parameters corresponding to the power battery to generate electrical simulation field, thermal simulation field and mechanical simulation field corresponding to the power battery;

[0014] Step S14: Perform three-dimensional mapping coupling construction on the topology connection architecture of each component of the power battery based on the electrical simulation field, thermal simulation field and mechanical simulation field corresponding to the power battery to generate a three-dimensional physical field coupling model of the power battery.

[0015] Further, step S2 includes the following steps:

[0016] Step S21: Extracting battery thermal parameters from the micro material scale of the power battery by the three-dimensional physical field coupling model of the power battery to obtain micro material thermal physical property data, including electrode material thermal conductivity and electrolyte specific heat capacity;

[0017] Step S22: Performing single cell charging and discharging thermal analysis from the battery monomer scale of the power battery by the three-dimensional physical field coupling model of the power battery to obtain battery monomer thermal characteristic data, including corresponding temperature change curve and internal resistance heat generation rate during charging and discharging process;

[0018] Step S23: Performing module thermal distribution analysis from the battery module scale of the power battery by the three-dimensional physical field coupling model of the power battery to obtain battery module thermal distribution data, including temperature difference between each monomer in the battery module and heat transfer path;

[0019] Step S24: Performing system thermal management statistical analysis from the thermal management system scale of the power battery by the three-dimensional physical field coupling model of the power battery to obtain battery system thermal management parameter data, including cooling medium flow and inlet temperature distribution;

[0020] Step S25: Merging micro material thermal physical property data, battery monomer thermal characteristic data, battery module thermal distribution data and battery system thermal management parameter data in the same data set to obtain power battery thermal characteristic data set.

[0021] Further, step S22 includes the following steps:

[0022] Performing monomer material tensor construction from the battery monomer scale of the power battery by the three-dimensional physical field coupling model of the power battery to analyze the microstructure changes of electrode and electrolyte materials at different temperatures and state of charge on the battery monomer, and abstract the conduction path of ions and electrons in the material into tensor form to construct electrode-electrolyte microstructure conductivity tensor, wherein the tensor elements represent the conductivity value of the corresponding material inside the guide;

[0023] Based on the electrode-electrolyte microstructure conductivity tensor and combined with the entropy change theory in thermodynamics, the ion migration network of the corresponding battery monomer in the three-dimensional physical field coupling model of the power battery is generated to study the ion migration behavior between the electrode and the electrolyte in the battery monomer during the charging and discharging process, and the ion migration behavior is regarded as the flux transmission driven by entropy change, taking the interface and internal pore of the electrode and electrolyte as nodes, and the ion migration path as edges, and by calculating the entropy change value of each node and the flux size of the edge, the corresponding ion migration entropy flux network is constructed;

[0024] The surface roughness and contact pressure distribution of the current collector and the electrode material are analyzed by considering the contact characteristics between the current collector and the electrode material in the corresponding battery cell in the three-dimensional physical field coupling model of the power battery and introducing the contact thermal resistance concept in heat transfer, and the thermal resistance of each contact area between the current collector and the electrode material is calculated according to the surface roughness and contact pressure distribution of the current collector and the electrode material, so as to construct the current collector-electrode contact thermal resistance matrix;

[0025] The charge and discharge heat source of the corresponding battery cell in the three-dimensional physical field coupling model of the power battery is analyzed based on the ion migration entropy flux network and the current collector-electrode contact thermal resistance matrix, so as to calculate the electrochemical reaction heat of the battery cell in the charging and discharging process according to the entropy change and reaction enthalpy change in the ion migration process, solve the Joule heat of the battery cell in the charging and discharging process through the electrode-electrolyte microstructure conductivity tensor, current distribution and thermal resistance, and couple the electrochemical reaction heat and the Joule heat in the three-dimensional space of the battery cell, so as to construct the electrochemical reaction heat-Joule heat coupled heat source body;

[0026] The non-steady-state heat transfer of the corresponding battery cell in the three-dimensional physical field coupling model of the power battery is analyzed based on the electrochemical reaction heat-Joule heat coupled heat source body and combined with the geometric shape of the battery cell and the micro material thermal physical property data, so as to solve the heat conduction temperature distribution through the finite difference method or the finite element method, so as to obtain the temperature distribution of the battery cell at different times and different positions in the charging and discharging process; the temperature distribution of the battery cell at different times and different positions in the charging and discharging process is fitted into a non-steady-state heat transfer three-dimensional surface, the points on the surface correspond to the spatial position and temperature distribution of the battery cell, and the change trend of the surface forms a temperature change curve, and the internal resistance heat generation rate is obtained by calculating the heat flux density gradient.

[0027] Further, the step S23 comprises the following steps:

[0028] The thermal temperature distribution of each cell corresponding to the battery module is obtained by analyzing the thermal temperature distribution of each cell corresponding to the battery module from the scale of the battery module corresponding to the power battery through the three-dimensional physical field coupling model of the power battery.

[0029] The temperature difference between each cell in the battery module is quantified according to the thermal temperature distribution of each cell in the battery module, so as to obtain the temperature difference between each cell in the battery module.

[0030] The module cell heat transfer between the battery module and the battery cell corresponding to the three-dimensional physical field coupling model of the power battery is analyzed based on the thermal temperature distribution of each cell corresponding to the battery module, so as to obtain the heat transfer path between each cell in the battery module.

[0031] Further, the module monomer heat transfer analysis between the corresponding battery modules and battery monomers in the power battery three-dimensional physical field coupling model based on the corresponding thermal temperature distribution of each monomer in the battery module includes the following steps:

[0032] Determine the contact interface between the battery monomer and the module structure through the corresponding battery module in the power battery three-dimensional physical field coupling model;

[0033] Obtain the material surface roughness measurement data and the contact pressure distribution through the corresponding battery module in the power battery three-dimensional physical field coupling model, and evaluate the heat flow distribution of the contact interface between the battery monomer and the module structure based on the material surface roughness measurement data and the contact pressure distribution, to obtain the heat flow distribution coefficient of the contact interface between the battery monomer and the module structure, wherein represents the proportion of heat flow from the battery monomer to the module structure at the contact interface;

[0034] Perform module monomer heat transfer analysis between the corresponding battery modules and battery monomers in the power battery three-dimensional physical field coupling model based on the corresponding thermal temperature distribution of each monomer in the battery module and the heat flow distribution coefficient of the contact interface between the battery monomer and the module structure, to obtain the heat transfer path between each monomer in the battery module.

[0035] Further, step S3 includes the following steps:

[0036] Step S31: Based on the neural network, a multi-scale deep neural network architecture is constructed combined with the power battery thermal characteristic data set, which includes an input layer, a hidden layer and an output layer;

[0037] Step S32: Based on the power battery thermal characteristic data set, the multi-scale deep neural network architecture is trained for thermal management prediction, so as to input the power battery thermal characteristic data set as a training set into the corresponding input layer of the multi-scale deep neural network architecture, and automatically assign weights of different power battery thermal characteristics corresponding features in the hidden layer, while designing a hybrid loss with an adaptive learning rate adjustment model parameter, wherein the hybrid loss includes temperature prediction error, heat generation rate prediction error and heat conduction coefficient prediction error, and the output layer is used to simultaneously predict the thermal management prediction parameters of the corresponding battery temperature distribution, heat generation efficiency and heat transfer coefficient of the power battery, so as to train and generate a power battery thermal management prediction model;

[0038] Step S33: Obtain the real-time battery thermal characteristic data set of the power battery, and input the real-time battery thermal characteristic data set of the power battery into the power battery thermal management prediction model for thermal management prediction, to output the corresponding thermal management prediction result of the power battery.

[0039] Further, step S31 includes the following steps:

[0040] The correlation between the thermal characteristic parameters of each power battery is obtained through the power battery thermal characteristic data set, and the number of nodes corresponding to the input layer and the feature combination mode are determined based on the correlation between the thermal characteristic parameters of each power battery, and a hierarchical feature extraction module is designed based on the neural network design, so that different feature extraction methods are adopted for different power battery thermal characteristic parameters, including convolutional neural network (CNN) for processing spatial features and recurrent neural network (RNN) for processing time series features;

[0041] The hidden layer is designed and the residual block and attention mechanism are introduced therein, wherein the residual block is used to solve the gradient disappearance of the deep network, and the attention mechanism is used to automatically assign the weight of different features;

[0042] The output layer corresponding to the multi-task learning is designed to simultaneously predict the thermal management prediction parameters of the battery temperature distribution, the heat generation efficiency and the heat transfer coefficient of the power battery, and the multi-scale deep neural network architecture is constructed through the input layer, the hidden layer and the output layer.

[0043] Further, step S4 comprises the following steps:

[0044] Step S41: Based on the thermal management prediction result of the power battery, the thermal management state change analysis of the thermal management system corresponding to the power battery is carried out, and the power battery thermal management state change space data is obtained;

[0045] Step S42: According to the power battery thermal management state change space data, the thermal management target function corresponding to the power battery is defined, which includes the optimization target functions of temperature distribution uniformity, energy consumption and power battery life;

[0046] Step S43: The thermal management constraint condition is designed by the thermal management system corresponding to the power battery, which includes the temperature upper and lower limit constraint, the cooling system power constraint and the charging and discharging current constraint, and the thermal management constraint optimization control is carried out on the thermal management target function corresponding to the power battery based on the thermal management constraint condition, so as to generate the optimal thermal management control strategy of the power battery;

[0047] Step S44: The optimal thermal management control strategy of the power battery is uploaded to the cloud platform and converted into actual thermal management system control parameters, including cooling pump speed, fan power and heating element current, which responds to control the corresponding thermal management system of the power battery to perform corresponding thermal management control work.

[0048] Further, the application also provides a neural network-based power battery thermal management simulation system for executing the neural network-based power battery thermal management simulation method as described above, which comprises:

[0049] A three-dimensional coupling simulation construction module is configured to acquire corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery through coupling experimental measurement, and to perform three-dimensional coupling simulation construction on the power battery based on the corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery, so as to generate a three-dimensional physical field coupling model of the power battery.

[0050] A battery thermal characteristic evaluation module is configured to perform battery thermal characteristic evaluation on the power battery according to micro material, battery monomer, battery module and thermal management system from four scales of the power battery, so as to obtain a power battery thermal characteristic data set, which includes micro material thermal physical property data, battery monomer thermal characteristic data, battery module thermal distribution data and battery system thermal management parameter data.

[0051] A thermal management prediction module is configured to construct a multi-scale deep neural network architecture based on a neural network, and to perform thermal management prediction training on the multi-scale deep neural network architecture based on the power battery thermal characteristic data set, so as to generate a power battery thermal management prediction model and output corresponding thermal management prediction results of the power battery, which include battery temperature distribution, thermal generation efficiency and thermal transfer coefficient corresponding thermal management prediction parameters.

[0052] A thermal management constraint control module is configured to perform thermal management constraint optimization control on the thermal management system corresponding to the power battery based on the corresponding thermal management prediction results of the power battery, so as to generate a power battery optimal thermal management control strategy, and to upload the power battery optimal thermal management control strategy to a cloud platform to respond to the thermal management system corresponding to the power battery to perform corresponding thermal management control work.

[0053] The beneficial effects of the present application are as follows:

[0054] 1、The power battery thermal management simulation method based on the neural network has the beneficial effects that compared with the prior art, through accurate experimental measurement, the performance parameters of the power battery in different fields are obtained, the parameters include electrical performance (such as voltage, current, charge and discharge efficiency), thermal performance (such as temperature, thermal conductivity) and mechanical performance (such as stress, deformation), they are the basis of constructing the battery comprehensive performance model, through coupling experimental measurement, various physical phenomena of the power battery can be comprehensively captured, and then on the basis of multi-physical field coupling, a three-dimensional physical field coupling model is constructed, the model has high accuracy, can reflect the comprehensive behavior of the battery under different environments and working conditions, through the coupling of three-dimensional physical fields, the mutual influence and dynamic change of multiple variables such as electricity, heat and force in the battery can be simulated in real time, which provides strong support for subsequent battery performance prediction and optimization, so that the interactive process of multi-physical field coupling such as electrochemical reaction, heat conduction and fluid flow can be accurately described, the comprehensive understanding of the multi-physical field characteristics of the battery is ensured, and accurate input data is provided for subsequent simulation and optimization control. Secondly, the thermal characteristics of the battery are evaluated in multiple scales through the three-dimensional physical field coupling model of the power battery, which involves multiple levels such as micro materials, battery monomers, battery modules and thermal management systems, the thermal property data of the micro materials help to understand the thermal conductivity and heat capacity of the battery; the thermal characteristic data of the battery monomer is the key to evaluate the thermal distribution in the monomer; the thermal distribution data of the battery module further reveals the thermal dynamic performance of the battery after assembly; and the thermal management system parameters are the core elements of optimizing the battery temperature control strategy, through multi-scale evaluation, the thermal behavior data of the battery in different levels can be accurately obtained, and the corresponding thermal characteristic data set is generated, these data sets will provide high-quality input data for the optimization of battery thermal management, and ensure the accuracy and reliability of the prediction results, the key of this step is to master the thermal characteristics of the battery comprehensively and accurately through fine hierarchical division and evaluation, so as to provide necessary data support for subsequent thermal management optimization and strategy formulation. Then, a multi-scale deep neural network architecture is constructed by using deep learning technology to train the collected battery thermal characteristic data set through a neural network model, the network architecture can efficiently predict the thermal management behavior of the battery by automatically learning the complex patterns in the data, through the training of the neural network, the accurate prediction of key thermal management parameters such as battery temperature distribution, heat generation efficiency and heat transfer coefficient can be realized, the advantage of the deep neural network lies in its powerful nonlinear modeling capability, which can process high-dimensional and complex data and extract potential thermal management rules from them, the key of this step is to quickly and accurately predict the thermal characteristics of the battery, so as to provide decision basis for the optimization of the battery thermal management system. At the same time, with the help of the learning ability of the neural network, the prediction model can be continuously optimized, the accuracy and robustness of the thermal management prediction can be improved, and the thermal management requirements of the battery under different working conditions can be adapted.Finally, the battery thermal management system is optimized and controlled based on the output results of the thermal management prediction model, so as to generate optimal thermal management control strategies by constraining and optimizing parameters such as the thermal distribution, heat generation efficiency and heat transfer coefficient of the battery, which can adjust the cooling and heating mode of the battery in real time, ensure that the battery is maintained within the optimal temperature range under various working conditions, and thus improve the performance and life of the battery. In addition, by uploading the optimal control strategy to the cloud platform, remote control and real-time monitoring of the thermal management system can be realized, ensuring that the thermal management system can automatically adjust the strategy according to real-time data to cope with different working environments and use conditions. The key of this step is to realize the intelligentization and automation of battery thermal management, to improve the overall performance and safety of the battery system through the optimization of the control strategy, and to improve the flexibility and control effect of the power battery thermal management through the intelligent management of the cloud platform.

[0055] 2、The power battery thermal management simulation system based on the neural network provided by the present application is composed of a three-dimensional coupling simulation construction module, a battery thermal characteristic evaluation module, a thermal management prediction module and a thermal management constraint control module, and can realize the power battery thermal management simulation method based on the neural network, which is used to realize the power battery thermal management simulation method based on the neural network by combining the operations between the computer programs running on each module. The internal structure of the system cooperates with each other, which can greatly reduce the repetitive work and labor input, and can quickly and effectively provide more accurate and efficient power battery thermal management simulation process based on the neural network, thereby simplifying the operation process of the power battery thermal management simulation system based on the neural network. BRIEF DESCRIPTION OF DRAWINGS

[0056] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments made with reference to the accompanying drawings:

[0057] Figure 1 The step flowchart of the power battery thermal management simulation method based on the neural network of the present application is shown in the figure.

[0058] Figure 2 The detailed step flowchart of step S1 in the figure. Figure 1 The detailed step flowchart of step S1 in the figure. DETAILED DESCRIPTION

[0059] The technical method of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.

[0060] To achieve the above object, please refer to Figures 1 to 2 The application provides a power battery thermal management simulation method based on a neural network, and in the embodiment of the application, please refer to Figure 1 The application provides a power battery thermal management simulation method based on a neural network, and in the embodiment of the application, please refer to

[0061] Step S1: Obtain the corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery through coupling experimental measurement, and perform three-dimensional coupling simulation construction on the power battery based on the corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery, to generate a three-dimensional physical field coupling model of the power battery;

[0062] In the embodiment of the application, the electrical, thermal and mechanical performance parameters of the power battery are obtained through coupling experimental measurement. For a certain type of ternary lithium battery, a high-precision battery test system is used to perform a charge-discharge experiment, and the electrical performance parameters such as voltage, current and capacity of the battery are measured at different charge-discharge rates such as 0.5C, 1C and 2C, with a sampling frequency of 1 Hz, and the data during the entire charge-discharge process is recorded continuously. An infrared thermal imager is used to measure the surface temperature distribution of the battery, and the temperature resolution of the thermal imager is 0.1℃ and the spatial resolution is 0.5mm. Temperature data is collected every 5 seconds during the charge-discharge process to obtain the dynamic changes of the surface temperature field of the battery. A mechanical testing device is used to apply pressure in different directions to the battery to measure the stress-strain curve and elastic modulus of the battery. In the pressure test, the loading rate is 0.1mm / min and the maximum loading force is 500N. The relationship data between pressure and deformation is recorded. Based on these measurement data, three-dimensional coupling simulation construction is performed on the power battery. A finite element analysis method is used to divide the battery into 100x80x50 hexahedral grid elements, each with a size of 1.5mmx1mmx0.4mm. An electrochemical-thermal-mechanical coupling model is established. The electrochemical model uses a P2D (pseudo two-dimensional) model to describe the electrochemical process inside the battery. The thermal model is based on the Fourier heat conduction equation to describe heat transfer. The mechanical model uses a linear elastic model to describe the mechanical response of the battery. The interaction between the three physical fields is realized through the coupling interface. For example, the Joule heat generated by the electrochemical process is input into the thermal model as a heat source. Temperature changes will affect the electrochemical performance parameters of the battery. At the same time, mechanical deformation will change the contact state inside the battery, thereby affecting the electrical and thermal performance. By solving the coupled equation set, a three-dimensional physical field coupling model of the power battery is finally generated. This model can accurately reflect the multi-physical field interaction of the battery in actual operation.

[0063] Step S2: obtaining the thermal characteristic data set of the power battery, including the thermal physical property data of the micro material, the thermal characteristic data of the battery monomer, the thermal distribution data of the battery module and the thermal management parameter data of the battery system, by the three-dimensional physical field coupling model of the power battery, according to the battery thermal characteristic evaluation from the four scales of the micro material, the battery monomer, the battery module and the thermal management system;

[0064] In the embodiment of the present application, by using the three-dimensional physical field coupling model of the power battery, the battery thermal characteristic evaluation is carried out from the four scales of the micro material, the battery monomer, the battery module and the thermal management system. In the micro material scale, the thermal physical property parameters of the electrode material and the electrolyte are analyzed by the model. For the electrode material, the interatomic potential energy and thermal vibration are calculated by the molecular dynamics simulation method, so as to obtain the thermal conductivity coefficient of the electrode material. The simulation system contains 1000 atoms, the simulation time step is 1 fs, the equilibrium simulation is carried out for 100 ps under the NVT ensemble, and then the production simulation is carried out for 500 ps under the NVE ensemble. The thermal conductivity coefficient is obtained by calculating the integral of the heat flow autocorrelation function. Finally, the thermal conductivity coefficient of the lithium nickel cobalt manganese oxide electrode material is determined as 1.2 W / (m K). For the electrolyte, the specific heat capacity is calculated by the statistical thermodynamics method. The molar fraction and heat capacity contribution of each component in the electrolyte are considered. The specific heat capacity of the lithium hexafluorophosphate solution is calculated as 1200 J / (kg K). In the battery monomer scale, the monomer charging and discharging thermal analysis is carried out. The charging and discharging working condition is set as 1C constant current charging to 4.2V, then constant voltage charging until the current decreases to 0.05C, and then 1C constant current discharging to 2.7V. The electrochemical reaction heat and Joule heat calculation module in the model is used to simulate the heat generation of the battery monomer in the charging and discharging process, combined with the constructed electrochemical reaction heat-Joule heat coupling heat source body. The temperature distribution inside the battery monomer is calculated by solving the non-steady-state heat conduction differential equation by the finite element method. In the charging process, the center temperature of the battery is recorded in real time. From the initial 25℃, the temperature gradually rises with the charging, and reaches 45℃ at 30 minutes. The complete temperature change curve is generated. At the same time, the internal resistance heat generation rate is calculated according to the heat flux density gradient. In the middle of the charging, the internal resistance heat generation rate reaches the peak value of 500 W / m 3, get the overall battery cell thermal characteristic data, on the battery module scale, conduct the module thermal distribution analysis, the module is composed of 12 battery cells in series, by determining the contact interface between the battery cell and the module structure (metal frame, heat sink), obtain the material surface roughness and contact pressure distribution data, calculate the heat flow distribution coefficient, simulate the heat transfer process of the module under the same charge and discharge conditions in the model, combined with the thermal temperature distribution data of each cell, use the Fourier law to calculate the heat flow size and direction between the cells and between the cells and the structure. For example, when charging for 30 minutes, part of the heat (60%) of the middle cells with higher temperature (such as cells 6 and 7) is conducted through the heat sink, and 40% is transmitted to the edge cells through the metal frame; there is also heat transfer between adjacent cells, such as between cell 1 and cell 2 according to the contact interface heat flow distribution coefficient (assuming 0.5). Through calculation and analysis, the temperature difference between each cell in the battery module is obtained, the maximum temperature difference is 6℃, and the detailed heat transfer path is determined, forming a complete battery module thermal distribution data, on the scale of the thermal management system, conduct statistical analysis of the system thermal management, the thermal management system adopts liquid cooling, and the cooling medium is glycol water solution, build a complete three-dimensional structure of the thermal management system in the model, including cooling pipeline, water pump, radiator and other components, set the initial parameters of the cooling medium, the water pump flow is 5 L / min, the inlet temperature is 20℃, simulate the process of heat generated by the battery module in the charging and discharging process being transferred to the cooling pipeline and exchanging heat with the cooling medium. Using the computational fluid dynamics (CFD) method, the fluid flow and heat transfer in the cooling pipeline are simulated, the cooling pipeline is divided into 500x100x50 grid units, the continuity equation, momentum equation and energy equation are solved, and the flow and temperature changes of the cooling medium at different positions in the pipeline are monitored in real time. In the area close to the heat source of the battery module, the temperature of the cooling medium increases obviously, and the outlet temperature reaches 28℃; at the same time, through the flow sensor simulation data, the flow distribution of different sections in the pipeline is obtained, the flow fluctuates slightly at the pipeline bend, but the overall average flow is stable at 5 L / min. These data are sorted and summarized to obtain the battery system thermal management parameter data including the cooling medium flow and inlet temperature distribution. The micro material thermal property data, battery cell thermal characteristic data, battery module thermal distribution data and battery system thermal management parameter data are integrated, stored in the same data set according to the unified data format and standard, and a complete power battery thermal characteristic data set is formed.

[0065] Step S3: based on the neural network combined with the power battery thermal characteristic data set, a multi-scale deep neural network architecture is constructed, and based on the power battery thermal characteristic data set, the multi-scale deep neural network architecture is trained for thermal management prediction, to generate a power battery thermal management prediction model, and output the corresponding thermal management prediction results of the power battery, including battery temperature distribution, heat generation efficiency and heat transfer coefficient corresponding thermal management prediction parameters;

[0066] In the embodiment of the present application, a multi-scale deep neural network architecture is constructed based on the neural network combined with the power battery thermal characteristic data set. First, the data set is preprocessed, including data cleaning, normalization and other operations, to remove outliers and noise points in the data set, for example, a certain temperature data point is obviously deviated from the normal range, which is excluded, and a minimum-maximum normalization method is used to map all feature data to the [0, 1] interval, the formula is , wherein is the original data, and respectively, the minimum and maximum values of the feature, a neural network architecture is constructed, the input layer is set to 40 neurons, corresponding to the 40 thermal characteristic features after preprocessing, a multi-layer structure is adopted for the hidden layer, 3 hidden layers are set, the number of neurons in each layer is 100, 80 and 60 respectively, the neurons in each hidden layer are connected to the neurons in the previous layer through weights, and full connection is adopted between layers, a residual block and an attention mechanism are introduced in the hidden layer, the residual block is used to solve the gradient vanishing problem of the deep network, and each residual unit contains two fully connected layers and a skip connection; the attention mechanism is used to automatically assign weights to different features, the input features are weighted by calculating the attention weights, the output layer is set to 3 neurons, corresponding to the prediction of the battery temperature distribution, the heat generation efficiency and the heat transfer coefficient, and the multi-scale deep neural network architecture is trained for thermal management prediction based on the power battery thermal characteristic dataset. The dataset is divided into a training set and a validation set in a ratio of 8:2, wherein the training set is used for model parameter adjustment, and the validation set is used for evaluating the model training effect. In the training process, the training set data is sequentially input into the input layer of the neural network, after the data enters the hidden layer, each neuron weights and sums the input data according to the connection weights, and performs nonlinear transformation through an activation function (such as a ReLU function). The weights of different power battery thermal characteristics corresponding to the features are automatically adjusted in the hidden layer through a back propagation algorithm, a hybrid loss function is designed, the temperature prediction error, the heat generation rate prediction error and the heat conduction coefficient prediction error are considered comprehensively, the mean square error (MSE) is used to calculate the error between the predicted values and the actual values of the three parameters, and then they are weighted and summed according to a certain proportion (such as temperature prediction error proportion 40%, heat generation rate prediction error proportion 30%, heat conduction coefficient prediction error proportion 30%), to obtain the final hybrid loss value. The adaptive learning rate adjustment algorithm (such as the Adam algorithm) is used to dynamically adjust the learning rate according to the change of the hybrid loss value. In the early stage of training, a larger learning rate can speed up the update speed of the model parameters; as the training progresses, when the loss value decreases to a certain extent, the learning rate is automatically reduced to avoid over updating of the model parameters, so as to ensure the stability and accuracy of the model training. After multiple rounds of training (the training period is set to 100 times), the prediction error of the model on the validation set gradually decreases and tends to be stable, and finally a power battery thermal management prediction model is trained and generated, and the thermal management prediction is performed by using the model. The real-time collected battery thermal characteristic data is input into the model, and the corresponding thermal management prediction results of the power battery are output after the calculation and processing of the model, including the thermal management prediction parameters corresponding to the battery temperature distribution, the heat generation efficiency and the heat transfer coefficient.

[0067] Step S4: based on the thermal management prediction result corresponding to the power battery, the thermal management system corresponding to the power battery is controlled for thermal management constraint optimization, to generate the optimal thermal management control strategy of the power battery; the optimal thermal management control strategy of the power battery is uploaded to the cloud platform to respond to the thermal management system corresponding to the power battery to perform corresponding thermal management control work.

[0068] In the embodiment of the application, by controlling the thermal management system for thermal management constraint optimization based on the thermal management prediction result corresponding to the power battery, taking the prediction result at a certain moment as an example, it is predicted that the battery module center temperature will rise from 45℃ to 48℃ and the edge cell temperature will rise from 42℃ to 44℃ within 10 minutes in the future, the heat generation efficiency is 500J per minute, and the heat transfer coefficient is 15 W / (m 2 K), by three-dimensional visualization technology, the prediction data is mapped into the three-dimensional model of the power battery thermal management system to generate thermal management state change space data, in the model, different colors represent different temperature regions, arrows represent heat flow direction, color depth and arrow thickness reflect temperature value and heat flow size, further analyze the change trend of temperature distribution, calculate the temperature gradient, under the current prediction condition, the temperature gradient between the center and the edge of the battery module is (48℃-44℃) / 0.1m=40K / m (assuming the module size is 0.1m), according to the power battery thermal management state change space data, define the thermal management objective function, the temperature distribution uniformity objective function adopts the standard deviation of each cell temperature in the battery module, the target is to minimize the standard deviation; the energy consumption objective function considers the power consumption of the cooling pump, fan and heating element, the target is to minimize the total energy consumption under the premise of meeting the temperature control requirement; the power battery life objective function is constructed according to the temperature and battery cycle life relationship model, the target is to keep the battery working temperature in the best range as much as possible to maximize the battery life, the three objective functions are combined by weighting to obtain the complete thermal management objective function , 、 and are weight coefficients, and , here =0.4、 =0.3、 =0.3, design thermal management constraints, set temperature upper and lower limit constraints, the upper limit of the temperature of the battery monomer is 55 DEG C, and the lower limit is 15 DEG C; cooling system power constraints, the maximum power of the cooling pump is 500W, the maximum power of the fan is 300W, and the maximum power of the heating element is 200W; charge and discharge current constraints, the maximum charging current is 2C (40A), and the maximum discharging current is 3C (60A), based on these constraints, the thermal management objective function is optimized and solved by using a sequential quadratic programming algorithm, and after multiple iterations, the optimal thermal management control parameters are finally obtained, including the optimal speed of the cooling pump, the optimal power of the fan and the optimal current of the heating element, the optimal thermal management control strategy of the power battery is generated, the generated optimal thermal management control strategy of the power battery is uploaded to the cloud platform and sent to the server of the cloud platform, after the cloud platform receives the control parameters, format verification and parameter rationality check are performed, after verification, the cloud platform converts the control parameters into actual thermal management system control signals, for the cooling pump, the speed parameter is converted into a corresponding voltage control signal, the supply voltage of the cooling pump is adjusted through the frequency converter to realize speed control, for the fan, the power parameter is converted into a PWM (pulse width modulation) control signal, the duty cycle of the PWM signal is adjusted to control the speed of the fan, thereby realizing power regulation, for the heating element, the current parameter is converted into a current control signal, the resistance in the circuit is adjusted or a constant current source is used to control the current size of the heating element. After the thermal management system receives these control signals, the corresponding thermal management control work is performed to ensure that the battery works in a safe and efficient temperature range.

[0069] Further, as an embodiment of the present application, referring to Figure 2 , it is a detailed step flow diagram of step S1 in Figure 1 , in this embodiment, step S1 includes the following steps:

[0070] Step S11: obtain the corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery through coupling experiment measurement;

[0071] In the embodiment of the present application, by coupling experimental measurement on a certain type of power battery in a professional power battery test laboratory, electrical performance parameters are obtained by using a high-precision battery tester, under constant current charging conditions, the battery is charged at a current of 1C (10A), the tester records the terminal voltage, the amount of charge, the charging and discharging efficiency and other data in real time, collects data every 100 milliseconds, and continues to charge until the battery is fully charged, to obtain complete charging voltage-time curve and capacity change data; under constant current discharge conditions, the battery is discharged at a current of 1C, the voltage decay, discharge capacity and other parameters during discharge are monitored, and discharge performance data is obtained; thermal performance parameter measurement is assisted by an infrared thermal imager and a temperature sensor, 10 high-precision thermocouple temperature sensors are uniformly arranged on the surface of the battery, the sensor probe is closely attached to the key parts such as the battery tab and the surface of the cell, temperature data is collected at a frequency of 5 times per second; at the same time, the infrared thermal imager is used to shoot the temperature distribution image of the battery surface at a frame rate of 1 frame per second, and the temperature change at different stages is recorded synchronously during the charging and discharging process of the battery, for example, the tab temperature reaches 45℃ and the middle temperature of the cell is 38℃ after 30 minutes of charging; mechanical performance parameter measurement uses a mechanical testing machine to apply pressure and vibration load in different directions to the battery, in the pressure test, the battery is applied with pressure at a loading rate of 10N per second, the pressure value is monitored in real time by using a pressure sensor, and the displacement sensor records the deformation amount of the battery, when the pressure reaches 5000N, the loading is stopped, and the pressure-deformation curve is obtained; in the vibration test, the vibration frequency is set to 20-200Hz and the amplitude is 2mm, the acceleration response data of the battery in the vibration process is collected by the acceleration sensor, and the mechanical performance parameters of the battery are comprehensively obtained.

[0072] Step S12: Obtain the design structure data corresponding to each component of the power battery, and perform topology connection construction on the power battery based on the design structure data corresponding to each component of the power battery, to generate a topology connection architecture of each component of the power battery.

[0073] In the embodiment of the present application, by extracting the design structure data corresponding to each component from the power battery design document, the ternary lithium battery includes positive plate, negative plate, separator, electrolyte, tab, shell and other components, using three-dimensional modeling software, according to the size parameters in the design drawing, the three-dimensional model of each component is accurately constructed in millimeter unit, for example, the length of the positive plate is set to 150mm, the width is 80mm, and the thickness is 0.1mm; the shell is a cuboid structure, the length is 200mm, the width is 100mm, and the height is 20mm, and the topological connection is constructed according to the actual assembly relationship of the battery, the positive plate and the negative plate are alternately stacked, and the separator is inserted in the middle, the thickness of the separator is 0.02mm, which plays a role of isolating the positive and negative electrodes to prevent short circuit; the tabs are welded on the positive plate and the negative plate respectively as the connecting components of current output and input; finally, the assembled battery cell is placed in the shell, the shell is made of aluminum alloy material, and is fixed and sealed by bolts, in the three-dimensional modeling software, the assembly constraint function is used to accurately set the position relationship and connection mode between each component, such as the welding fixation of the tab and the positive and negative plates, and the wrapping assembly of the shell and the internal battery cell, to generate the complete topological connection architecture of each component of the power battery, and clearly present the spatial layout and connection relationship of each component inside the battery.

[0074] Step S13: performing physical field simulation analysis on the power battery according to the electrical performance parameters, thermal performance parameters and mechanical performance parameters corresponding to the power battery, to generate electrical simulation field, thermal simulation field and mechanical simulation field corresponding to the power battery;

[0075] In the embodiment of the present application, by using professional multi-physical field simulation software to perform physical field simulation analysis on the power battery based on the obtained electrical performance parameters, thermal performance parameters and mechanical performance parameters, in the electrical simulation field construction, the current, voltage and other electrical performance parameters in the charging and discharging process are input into the software as boundary conditions, and the electrical properties parameters of the electrolyte, electrode material and other components inside the battery are set, such as the electrolyte conductivity of 0.01S / cm and the electrode material resistivity of 0.001Ω m, the software calculates the current density distribution and potential distribution inside the battery based on the finite element analysis method, generates the electrical simulation field of the battery in the charging and discharging process, and intuitively displays the flow path of the current inside the battery and the potential difference of each part; in the thermal simulation field construction, the temperature data collected by the temperature sensor and the temperature distribution obtained by the infrared thermal imager are used as initial conditions, and the thermal conductivity, specific heat capacity and other thermal property parameters of each component of the battery are set, such as the thermal conductivity of the shell of 200W / (m K) and the specific heat capacity of the battery cell of 800J / (kg K), the software simulates the heat generation, conduction and dissipation process of the battery in the charging and discharging process by solving the heat conduction equation, generates a thermal simulation field, and presents the trend of the internal temperature of the battery changing with time and the spatial distribution state. In the process of constructing the mechanical simulation field, the pressure-deformation data of the pressure test and the acceleration response data of the vibration test are taken as the load conditions, and the mechanical property parameters of each component of the battery, such as the elastic modulus and Poisson's ratio of the shell, are set, for example, the elastic modulus of the shell is 70GPa, and Poisson's ratio is 0.3. The software uses a mechanical analysis algorithm to calculate the stress and strain distribution of the battery under the action of pressure and vibration, generates a mechanical simulation field, and displays the stress state and deformation of each part of the battery under the mechanical load. Finally, the corresponding electrical simulation field, thermal simulation field and mechanical simulation field of the power battery are generated.

[0076] Step S14: based on the electrical simulation field, the thermal simulation field and the mechanical simulation field corresponding to the power battery, the topological connection architecture of each component of the power battery is three-dimensionally mapped and coupled to construct a three-dimensional physical field coupling model of the power battery.

[0077] In the embodiment of the present application, by three-dimensionally mapping and coupling the topological connection architecture of each component of the power battery based on the generated electrical simulation field, thermal simulation field and mechanical simulation field in the multi-physical field simulation software, the current density and potential distribution data in the electrical simulation field, the temperature distribution data in the thermal simulation field, and the stress and strain distribution data in the mechanical simulation field are mapped onto the three-dimensional model of the topological connection architecture of each component of the power battery according to the spatial positional relationship of each component, for example, in the process of battery charging and discharging, the electrical simulation field shows that the current density at the tab is large, and the current density distribution is mapped to the tab part of the three-dimensional model; the thermal simulation field shows that the temperature in the middle of the cell is high, and the temperature distribution is mapped to the corresponding position of the cell; the mechanical simulation field shows that the stress is concentrated at the corner of the shell, and the stress distribution is mapped to the corner area of the shell. Through data interaction and coupling algorithm, the software considers the mutual influence between the physical fields. When the battery is working, the current will generate heat, which will affect the thermal field distribution; the temperature change will affect the electrical and mechanical properties of the battery material, and then change the electrical and mechanical fields. Through repeated iterative calculation, data fusion and collaborative calculation between the three physical fields are realized, and finally a complete three-dimensional physical field coupling model of the power battery is generated. The model can intuitively and comprehensively show the interaction relationship and comprehensive performance between the electrical, thermal and mechanical properties of the battery in the working state.

[0078] Further, step S2 includes the following steps:

[0079] Step S21: extracting battery thermal parameters from the three-dimensional physical field coupling model of the power battery according to the micro material scale corresponding to the power battery to obtain micro material thermal physical property data, including electrode material thermal conductivity and electrolyte specific heat capacity;

[0080] In the embodiment of the present application, by focusing on the micro material scale in the power battery three-dimensional physical field coupling model, the battery thermal parameter extraction is carried out, taking a ternary lithium battery as an example, the electrode material of which is lithium nickel cobalt manganese oxide (NCM), and the electrolyte is lithium hexafluorophosphate (LiPF6) solution. The test equipment and simulation software in the field of material science are used for analysis combined with the microstructure data in the model. For the extraction of the thermal conductivity of the electrode material, the hot-wire method is used for measurement. A small hot-wire (diameter of 0.1 mm platinum wire) is buried in the prepared electrode material sample. A stable voltage is applied to the two ends of the sample, so that the hot-wire generates constant heat. The temperature changes at different positions around the hot-wire are measured by a high-precision temperature sensor (accuracy ±0.01℃). According to the Fourier heat conduction law and the calculation formula of the hot-wire method, the average value is obtained through multiple measurements. The thermal conductivity of the lithium nickel cobalt manganese oxide electrode material at 25℃ is 1.2 W / (m K). For the determination of the specific heat capacity of the electrolyte, a differential scanning calorimeter (DSC) is used. A certain amount of lithium hexafluorophosphate solution is placed in the DSC crucible. The temperature is raised from 10℃ to 50℃ at a constant heating rate (5℃ / min). At the same time, the empty crucible is measured as a reference under the same conditions. The instrument records the heat flow difference between the sample and the reference in real time. According to the relationship curve of heat flow difference and temperature, combined with the sample mass, the average specific heat capacity of the electrolyte in the temperature range is calculated as 1200 J / (kg K). These data are sorted and recorded, and finally the complete micro material thermal physical property data are obtained.

[0081] Step S22: performing monomer charging and discharging thermal analysis on the battery monomer scale according to the power battery three-dimensional physical field coupling model to obtain battery monomer thermal characteristic data, including corresponding temperature change curve and internal resistance heat generation rate in the charging and discharging process;

[0082] In the embodiment of the present application, based on the power battery three-dimensional physical field coupling model, monomer charging and discharging thermal analysis is carried out for the battery monomer scale. Taking the ternary lithium battery monomer as an example, the rated capacity is 20 Ah, and the nominal voltage is 3.7 V. The charging and discharging conditions are set in the model. The charging process adopts constant current and constant voltage charging mode. First, constant current charging is carried out at 1C (20A) current to 4.2V, and then constant voltage charging is carried out until the current decreases to 0.05C. The discharging process is constant current discharging at 1C current to 2.7V. The electrochemical reaction heat and Joule heat calculation module in the model is used, combined with the constructed electrochemical reaction heat-Joule heat coupling heat source body, to simulate the heat generation of the battery monomer in the charging and discharging process. The non-steady-state heat conduction differential equation is solved by the finite element method. The battery monomer is divided into 100×80×20 grid units. The initial temperature is set to 25℃, and the boundary condition is the convective heat transfer coefficient of 10 W / (m 2 K), ambient temperature 20℃. During the charging and discharging process, the temperature change data of the battery monomer at different positions (such as the center, edge, and tab) are recorded in real time, and collected every 10 seconds. After the charging starts, the battery monomer temperature gradually rises with the electrochemical reaction and the heat generated by the current passing through, and the center temperature reaches 45℃ at 30 minutes, and a complete temperature change curve is generated. At the same time, the internal resistance heat generation rate is calculated according to the heat flux density gradient, and the internal resistance heat generation rate reaches a peak value, for example, 500 W / m 3 , and finally the overall battery monomer thermal characteristic data is obtained.

[0083] Step S23: performing module thermal distribution analysis on the battery module scale corresponding to the power battery through the power battery three-dimensional physical field coupling model to obtain battery module thermal distribution data, including temperature difference between each monomer in the battery module and heat transfer path;

[0084] In the embodiment of the present application, the module thermal distribution analysis is performed on the battery module scale in the power battery three-dimensional physical field coupling model, and the module is composed of 12 battery monomers connected in series. Through corresponding operations, the contact interface between the battery monomer and the module structure (metal frame, heat sink) is determined, the material surface roughness and contact pressure distribution data are obtained, the heat flow distribution coefficient is calculated, the heat transfer process of the module under the same charging and discharging condition is simulated in the model, the heat flow size and direction between the monomers and between the monomer and the structure are calculated by using the Fourier law combined with the thermal temperature distribution data of each monomer, for example, part of the heat (60%) of the middle monomers (such as monomers 6 and 7) with higher temperature is conducted out through the heat sink, and 40% is transferred to the edge monomers through the metal frame; there is also heat transfer between adjacent monomers, such as monomer 1 and monomer 2, which exchange heat according to the contact interface heat flow distribution coefficient (assuming 0.5). Through calculation and analysis, the temperature difference between each monomer in the battery module is obtained, and the maximum temperature difference is 6℃. At the same time, the detailed heat transfer path is determined. The temperature distribution and heat transfer in the module are displayed with different colors and arrows by using three-dimensional visualization software, and complete battery module thermal distribution data is formed, which provides a basis for module thermal management design.

[0085] Step S24: performing system thermal management statistical analysis on the thermal management system scale corresponding to the power battery through the power battery three-dimensional physical field coupling model to obtain battery system thermal management parameter data, including cooling medium flow and inlet temperature distribution;

[0086] In the embodiment of the present application, by the corresponding thermal management system of the power battery, the system thermal management statistical analysis is carried out through the three-dimensional physical field coupling model of the power battery. The thermal management system adopts a liquid cooling mode, and the cooling medium is a glycol water solution. A complete three-dimensional structure of the thermal management system is constructed in the model, including cooling pipes, water pumps, radiators and other components. The initial parameters of the cooling medium are set. The water pump flow is 5 L / min, and the inlet temperature is 20℃. The heat generated by the battery module during the charging and discharging process is simulated to be transferred to the cooling pipe and to exchange heat with the cooling medium. By using the computational fluid dynamics (CFD) method, the fluid flow and heat transfer in the cooling pipe are simulated. The cooling pipe is divided into 500x100x50 grid units. The continuity equation, momentum equation and energy equation are solved. The flow and temperature changes of the cooling medium at different positions in the pipe are monitored in real time. In the area close to the battery module heat source, the cooling medium temperature rises obviously, and the outlet temperature reaches 28℃. At the same time, through the flow sensor simulation data, the flow distribution of different cross sections in the pipe is obtained. The flow fluctuates slightly at the pipe bend, but the overall average flow is stable at 5 L / min. These data are sorted and summarized to finally obtain the battery system thermal management parameter data including the cooling medium flow and inlet temperature distribution.

[0087] Step S25: merging the micro material thermal physical property data, the battery monomer thermal characteristic data, the battery module thermal distribution data and the battery system thermal management parameter data in the same data set to obtain the power battery thermal characteristic data set.

[0088] In the embodiment of the present application, by integrating the micro material thermal physical property data (electrode material thermal conductivity 1.2 W / (m K), electrolyte specific heat capacity 1200 J / (kg K) ), battery monomer thermal characteristic data (charging and discharging temperature change curve, internal resistance heat generation rate data), battery module thermal distribution data (temperature difference between monomers 6℃, heat transfer path information) and battery system thermal management parameter data (cooling medium flow 5 L / min, inlet temperature distribution, etc.), according to the unified data format and standard, the various data are stored in the same data set in the form of a table. Each row represents a data record, and each column corresponds to different types of data, such as the first column recording the data type identifier (micro material, monomer, module, system), and the subsequent columns storing specific thermal parameters, curve data, distribution information, etc. In this way, different scales and different types of thermal related data are organically combined to form a complete power battery thermal characteristic data set, which provides comprehensive data support for the power battery thermal management simulation based on neural network and the subsequent thermal management strategy optimization.

[0089] Further, step S22 includes the following steps:

[0090] The monomer material tensor is constructed on the basis of the generated three-dimensional physical field coupling model of the power battery, so as to analyze the microstructure changes of the electrode and electrolyte materials on the battery monomer under different temperatures and state of charge, and the conduction paths of ions and electrons in the materials are abstracted into tensor form to construct the electrode-electrolyte microstructure conductivity tensor, wherein the tensor elements represent the conductivity values in the corresponding material interior;

[0091] In the embodiment of the present application, the monomer material tensor is constructed on the basis of the generated three-dimensional physical field coupling model of the power battery, so as to analyze the microstructure changes of the electrode and electrolyte materials on the battery monomer under different temperatures and state of charge, and the conduction paths of ions and electrons in the materials are abstracted into tensor form to construct the electrode-electrolyte microstructure conductivity tensor, wherein the tensor elements represent the conductivity values in the corresponding material interior;

[0092] Preferably, the ion migration network of the corresponding battery monomer in the three-dimensional physical field coupling model of the power battery is generated based on the electrode-electrolyte microstructure conductivity tensor and combined with the entropy change theory in thermodynamics, so as to study the migration behavior of ions between the electrode and the electrolyte in the battery monomer during the charging and discharging process, and the migration behavior is regarded as the flux transmission driven by entropy change, the interface and internal pores of the electrode and the electrolyte are regarded as nodes, and the ion migration path is regarded as an edge, and the entropy change value of each node and the flux size of the edge are calculated, so as to construct the corresponding ion migration entropy change flux network;

[0093] In the embodiment of the present application, by means of the constructed electrode-electrolyte microstructure conductivity tensor, the ion migration network of the battery monomer in the three-dimensional physical field coupling model of the power battery is generated in combination with the entropy change theory in thermodynamics. Taking the charging process of the battery monomer as an example, the interface and internal pores of the electrode and electrolyte are set as nodes, and the ion migration path is set as an edge. The ion migration behavior in the battery monomer is simulated and analyzed by using the computational fluid dynamics software and the thermodynamic calculation module. In the calculation process, the ion migration behavior is regarded as the flux transmission driven by entropy change. For each node, the entropy change value is calculated by using the thermodynamic formula according to the temperature, ion concentration and other parameters at the node. For example, when the temperature is 30℃ and the lithium ion concentration changes at a certain node of the electrode-electrolyte interface, the entropy change value is calculated by the formula (wherein is the amount of substance, is the gas constant, , is the ion concentration at different times). For the edge (ion migration path), the flux size is calculated according to the ion migration rate and the amount of substance of the migrated ions. For example, if the lithium ion migrates 0.01 mol per second on a certain ion migration path, the flux is 0.01 mol / s. By calculating all the nodes and edges in the battery monomer, the corresponding ion migration entropy change flux network is constructed. The network directly presents the ion migration behavior between the electrode and the electrolyte in the battery monomer during the charging and discharging process, and the driving effect of entropy change on ion migration, which helps to deeply study the ion transmission mechanism in the battery and provides a theoretical basis for optimizing the battery performance.

[0094] Preferably, the surface roughness and contact pressure distribution of the current collector and the electrode material are analyzed by considering the contact characteristics between the current collector and the electrode material in the corresponding battery monomer in the three-dimensional physical field coupling model of the power battery and introducing the contact thermal resistance concept in heat transfer, and the thermal resistance size of each contact area between the current collector and the electrode material is calculated according to the surface roughness and contact pressure distribution of the current collector and the electrode material, so as to construct the current collector-electrode contact thermal resistance matrix.

[0095] In the embodiment of the present application, the contact thermal resistance concept in heat transfer is introduced for the contact characteristics between the current collector and the electrode material in the battery monomer in the three-dimensional physical field coupling model of the power battery, the surface profilometer is used to measure the surface of the current collector (such as aluminum foil) and the electrode material (NCM), and the surface roughness data are obtained, for example, the surface roughness Ra value of the current collector is 0.5 μm, and the surface roughness Ra value of the electrode material is 1.2 μm. At the same time, the contact pressure distribution between the current collector and the electrode material is measured by the pressure sensor array, and the contact pressure in different areas is recorded during the battery assembly process, for example, the contact pressure in the center area of the battery is 1.5 MPa, and the contact pressure in the edge area is 1.2 MPa. According to the calculation formula of the contact thermal resistance in heat transfer (wherein is the contact gap, is the thermal conductivity, is the contact area), the thermal resistance of each contact area between the current collector and the electrode material is calculated in combination with the surface roughness and the contact pressure distribution data. In the area where the contact pressure is large and the surface roughness is small, the contact gap is small, and the calculated value of the thermal resistance is 0.05 K / W; in the area where the contact pressure is small and the surface roughness is large, the calculated value of the thermal resistance is 0.2 K / W. The thermal resistances of the contact areas are arranged in a matrix form to construct a current collector-electrode contact thermal resistance matrix, which records the thermal resistances of different positions between the current collector and the electrode material in detail, provides key parameters for accurately simulating the heat conduction process in the battery, helps to optimize the design of the battery thermal management system, and improves the heat dissipation efficiency and safety of the battery.

[0096] Preferably, the corresponding battery monomer in the three-dimensional physical field coupling model of the power battery is analyzed based on the ion migration entropy flux network and the current collector-electrode contact thermal resistance matrix to calculate the electrochemical reaction heat of the battery monomer in the charging and discharging process according to the entropy change and reaction enthalpy change in the ion migration process. The Joule heat of the battery monomer in the charging and discharging process is solved through the electrode-electrolyte microstructure conductivity tensor, the current distribution and the thermal resistance, and the electrochemical reaction heat and the Joule heat are coupled in the three-dimensional space of the battery monomer to construct an electrochemical reaction heat-Joule heat coupling heat source body;

[0097] In the embodiment of the present application, the charging and discharging heat source analysis is carried out for a certain type of power battery monomer in the three-dimensional physical field coupling model of the power battery, the entropy change data of the ion migration of the battery monomer in the charging process are obtained based on the ion migration entropy flux network which has been constructed, for example, the entropy change value of the lithium ion migration is calculated as 8 J / (mol·K) at a certain electrode-electrolyte interface node according to the formula , and the reaction enthalpy change of the process is determined in combination with the chemical reaction formula. Taking the charging process of a ternary lithium battery as an example, the electrochemical reaction formula of the positive electrode is: The reaction formula of the negative electrode is: To determine the reaction enthalpy change, first, the standard molar enthalpy of formation data of the reactants involved need to be obtained, which can be found in authoritative thermodynamic data manuals, such as 、 、 and and other substances under standard conditions (298.15 K, 100 kPa) According to the calculation formula of reaction enthalpy change , is the stoichiometric number of the th product in the chemical reaction formula, is the stoichiometric number of the th reactant in the chemical reaction formula, and the electrochemical reaction heat at this node is calculated by multiplying the entropy change value and the reaction enthalpy change. The electrochemical reaction heat of all nodes in the battery monomer is accumulated to obtain the total electrochemical reaction heat of the entire battery monomer. Using the electrode-electrolyte microstructure conductivity tensor and the current distribution, combined with the current collector-electrode contact thermal resistance matrix, the Joule heat is calculated. The conductivity value of the electrode material along the crystal lattice direction is 100 S / m, and the vertical direction is 20 S / m. According to the current density and conductivity, the formula Q = I 2 R (where I is the current and R is the resistance calculated from conductivity and geometric size) is used to calculate the Joule heat inside the electrode material. The thermal resistance of the contact area between the current collector and the electrode material is considered, such as the thermal resistance of the area with high contact pressure being 0.05 K / W. The heat generated by the current flowing through the contact area is calculated by the formula ( is the contact voltage), and the Joule heat of the contact area is calculated. The Joule heat inside the electrode material and the contact area is summarized. Finally, the electrochemical reaction heat and the Joule heat calculated are coupled in the three-dimensional space of the battery monomer. Based on the three-dimensional model of the battery monomer, the electrochemical reaction heat and the Joule heat are distributed to the corresponding three-dimensional space grid in the form of heat flux according to their location and size, and the electrochemical reaction heat-Joule heat coupling heat source body is constructed. This heat source body accurately reflects the heat source distribution of the battery monomer at different positions during charging and discharging.

[0098] Preferably, based on the electrochemical reaction heat-Joule heat coupling heat source and combined with the geometry of the battery cell and the microscopic material thermal property data, an unsteady-state heat transfer analysis is performed on the corresponding battery cell in the three-dimensional physical field coupling model of the power battery. The thermal conductivity temperature distribution is solved by the finite difference method or the finite element method to obtain the temperature distribution of the battery cell at different times and locations during the charging and discharging process. The temperature distribution of the battery cell at different times and locations during the charging and discharging process is fitted into an unsteady-state heat transfer three-dimensional surface. The points on this surface correspond to the spatial location and temperature distribution of the battery cell. The change trend of the surface forms a temperature change curve. At the same time, the internal resistance heating rate is obtained by calculating the heat flux density gradient.

[0099] In this embodiment of the invention, an unsteady-state heat transfer analysis is performed on a single battery cell within a three-dimensional physical field coupling model of a power battery based on a constructed electrochemical reaction heat-Joule heat coupled heat source. The battery cell is known to be cuboid in shape, 150mm long, 80mm wide, and 20mm high. Furthermore, the thermal properties of the micromaterials, such as the positive electrode material, negative electrode material, and electrolyte, are obtained. For example, the thermal conductivity of the positive electrode material is 2W / ( ), specific heat capacity is 800 J / ( The finite element method was used to analyze the heat transfer of a single battery cell. The three-dimensional model of the battery cell was divided into 100×60×20 small mesh elements, and each element was assigned corresponding thermophysical parameters. Based on the unsteady-state heat conduction differential equation... (in For density, For specific heat capacity, For temperature, For time, Thermal conductivity, (Heat source intensity), the heat source intensity of the electrochemical reaction heat-Joule heat coupled heat source body. Substituting into the equation, the initial conditions are set as follows: initial temperature of the battery cell is 25℃, and the boundary conditions are as follows: convective heat transfer coefficient between the battery cell surface and the environment is 10W / ( With an ambient temperature of 20℃, finite element method (FEM) software was used for iterative solving to obtain temperature distribution data of individual battery cells at different times (e.g., 10 min, 30 min charging, 20 min discharging) and locations during the charging and discharging process. This temperature distribution data was then imported into 3D modeling software and fitted into an unsteady-state heat transfer 3D surface. Each point on the surface corresponds to a specific location and temperature value in the 3D space of the battery cell. For example, at 30 min charging, the temperature at the center of the battery cell on the surface is 45℃. The changing trend of the surface forms a temperature change curve, visually demonstrating the temperature variation of the battery cell over time and space. Simultaneously, by calculating the heat flux density gradient and using the formula... ( The heat generation rate of the internal resistance of the battery monomer is further obtained, and comprehensive data support is provided for the design and optimization of the power battery thermal management system.

[0100] Further, the step S23 comprises the following steps:

[0101] The corresponding thermal temperature distribution of each monomer in the battery module is obtained by analyzing the thermal temperature distribution of each monomer in the battery module according to the corresponding thermal temperature distribution of each monomer in the battery module based on the three-dimensional physical field coupling model of the power battery.

[0102] In the embodiment of the present application, by focusing on a power battery module composed of 12 battery monomers in the established three-dimensional physical field coupling model of the power battery, the temperature information of each monomer is mapped to the overall model of the module based on the previously obtained temperature distribution data of the battery monomer at different times and different positions during charging and discharging. For example, at the time of charging for 30 minutes, the center temperature of monomer 1 is 45℃, the center temperature of monomer 2 is 43℃, and so on. The temperature distribution data of the 12 monomers in the three-dimensional space is integrated, and the battery module is divided into 100x80x50 small spatial grid units by using professional three-dimensional modeling and analysis software through grid division technology. The size of each grid unit is 1.5mmx1mmx0.4mm. The temperature data of each monomer is accurately filled into the corresponding grid unit according to its actual position in the module, and the thermal temperature distribution cloud chart of the entire battery module at this time is generated. In the cloud chart, different colors represent different temperature intervals, the red area represents the high temperature area, and the blue area represents the low temperature area. By rotating, scaling and other operations, the temperature distribution of each monomer in the module can be observed from different angles, and the distribution characteristics such as the relatively low temperature of the monomers at the edge of the module and the high temperature of the monomers in the middle part are clearly presented. The corresponding thermal temperature distribution of each monomer in the battery module at the time of charging for 30 minutes is finally obtained. During discharging and at other times, the temperature distribution analysis is also carried out in the same way to obtain complete thermal temperature distribution data of the module.

[0103] Preferably, the temperature difference between each monomer in the battery module is quantified according to the corresponding thermal temperature distribution of each monomer in the battery module.

[0104] In the embodiment of the present application, the temperature difference is quantified by the previously obtained thermal temperature distribution corresponding to each monomer in the battery module, the temperature distribution data at the time of charging for 30 minutes is selected, the temperature at the center position of the battery monomer is taken as the representative value, the monomer with the highest temperature is found first, assuming that the center temperature of monomer 7 is 48℃, the monomer with the lowest temperature is found, assuming that the center temperature of monomer 3 is 42℃, the temperature difference between the two is calculated as 48℃-42℃=6℃, which is the maximum temperature difference between the monomers in the module. In order to more comprehensively describe the temperature difference, the standard deviation of all monomer temperatures is calculated, the center temperature values of 12 monomers, 45℃, 43℃, 42℃, 44℃, 46℃, 47℃, 48℃, 45℃, 44℃, 43℃, 42℃, 46℃, are substituted into the standard deviation calculation formula (wherein is the temperature value of each monomer, is the average value of all monomer temperatures, is the number of monomers), the average value is calculated first, then the square of the difference between each monomer temperature and the average value is calculated in turn, the sum is divided by the number of monomers, and finally the square root is taken to obtain a standard deviation of about 1.8℃. Through the maximum temperature difference and the standard deviation, the temperature difference between each monomer in the battery module is accurately obtained, which provides data support for evaluating the thermal consistency of the battery module.

[0105] Preferably, based on the thermal temperature distribution corresponding to each monomer in the battery module, the battery module and the battery monomer in the three-dimensional physical field coupling model of the power battery are analyzed for module monomer heat transfer to obtain the heat transfer path between each monomer in the battery module.

[0106] In the embodiment of the present application, based on the thermal temperature distribution corresponding to each monomer in the battery module, the battery module and the battery monomer in the three-dimensional physical field coupling model of the power battery are analyzed for module monomer heat transfer, taking the temperature distribution state at the time of charging for 30 minutes as an example, according to Fourier's law (wherein is the heat flux density, is the thermal conductivity, is the temperature gradient), the heat flux density between each monomer is calculated, assuming that monomer 1 and monomer 2 are adjacent, the temperature of monomer 1 is 45℃, the temperature of monomer 2 is 43℃, the contact area between the two monomers is 80cm 2 , the thermal conductivity k of the monomer material is 2W / ( ), the temperature gradient =(45-43)×0.01m=200K / m (assuming the center distance between the two monomers is 1cm), then the heat flux density =-400W / , the negative sign indicates that heat flows from the high-temperature monomer 1 to the low-temperature monomer 2, and the heat flow =-3.2W, by calculating the heat flow between all adjacent cells in the module, the size and direction of the heat flow between each two cells are determined, and the three-dimensional visualization software is used to mark the heat flow direction on the three-dimensional model of the battery module in the form of arrows, the thickness of the arrow represents the size of the heat flow, thereby intuitively showing the heat transfer path between each cell in the battery module, it can be clearly seen that in the module, heat is mainly transferred from the middle cell with higher temperature to the edge cell with lower temperature, and there is also mutual heat transfer between adjacent cells, and the complete heat transfer path between each cell in the battery module is obtained, which provides an important basis for optimizing the thermal management design of the battery module.

[0107] Further, the module cell heat transfer analysis between the corresponding battery module and the battery cell in the power battery three-dimensional physical field coupling model based on the corresponding thermal temperature distribution of each cell in the battery module includes the following steps:

[0108] The contact interface between the battery cell and the module structure is determined between the corresponding battery module and the battery cell in the power battery three-dimensional physical field coupling model;

[0109] In the embodiment of the application, the contact interface between the battery cell and the module structure is determined in the power battery three-dimensional physical field coupling model for a battery module composed of 12 battery cells, the module structure includes a metal frame, a heat sink and the like, wherein the metal frame is used to fix the battery cell, and the heat sink is responsible for leading out the heat generated by the battery, taking one cuboid battery cell as an example, among the six surfaces of the battery cell, four sides are closely attached to the metal frame, the bottom surface is in contact with the heat sink, and the top surface is in contact with the bottom surface of the adjacent cell, the contact areas are accurately identified and marked through the geometric analysis function of the three-dimensional modeling software, the three-dimensional models of the battery cell and the structure are subjected to Boolean operation by the software, and the intersecting surface domains are found, which are the contact interfaces, for example, the four sides of the battery cell in contact with the metal frame are accurately defined, and the corresponding contact interface geometric data including area, shape and the like are generated, thereby providing basic data for subsequent heat transfer analysis.

[0110] Preferably, the corresponding material surface roughness measurement data and the contact pressure distribution are obtained from the corresponding battery module in the power battery three-dimensional physical field coupling model, and the heat flow distribution of the contact interface between the battery cell and the module structure is evaluated based on the material surface roughness measurement data and the contact pressure distribution, so as to obtain the heat flow distribution coefficient of the contact interface between the battery cell and the module structure, wherein represents the proportion of the heat flow from the battery cell to the module structure at the contact interface;

[0111] In the embodiment of the present application, the material surface roughness measurement data and the contact pressure distribution corresponding to the battery module are obtained through the three-dimensional physical field coupling model, so as to obtain the surface roughness data by measuring the surfaces of the battery monomer, the metal frame and the heat dissipation plate using a surface profiler, such as the battery monomer surface roughness Ra value of 1.2 μm, the metal frame surface Ra value of 0.8 μm, and the heat dissipation plate surface Ra value of 0.6 μm. In the module assembly process, the contact pressure distribution is measured by using a pressure sensor array. In the contact area between the battery monomer and the metal frame, the central part contact pressure is 1.8 MPa, and the edge part is 1.5 MPa. In the contact area between the battery monomer and the heat dissipation plate, the average contact pressure is 2.0 MPa. Based on these data, the heat flow distribution of the contact interface between the battery monomer and the module structure is evaluated, so as to calculate the contact thermal resistance of each contact interface according to the contact thermal resistance theory, the contact thermal resistance is related to the surface roughness and the contact pressure, and the formula = f(Ra, p) (wherein is the contact thermal resistance, is the surface roughness, p is the contact pressure) is used. For example, in the contact area between the battery monomer and the metal frame, the contact thermal resistance is calculated to be 0.1 K / W according to the surface roughness and the contact pressure. The contact thermal resistance of the contact area with the heat dissipation plate is 0.05 K / W. Then, the heat flow distribution coefficient is calculated according to the thermal resistance. Assuming that the total heat flow generated by the battery monomer is , the heat flow transferred through the heat dissipation plate is , and the heat flow transferred through the metal frame is , the heat flow distribution coefficient ( is the contact thermal resistance of the battery monomer and the heat dissipation plate, is the contact thermal resistance of the battery monomer and the metal frame), for example, the heat flow distribution coefficient of the contact interface between the battery monomer and the heat dissipation plate is calculated to be 0.6, that is, the proportion of the heat flow from the battery monomer to the heat dissipation plate at the contact interface is 60%. The heat flow distribution coefficient of the contact interface with the metal frame is 0.4, so that the complete heat flow distribution coefficient is obtained.

[0112] Preferably, the module monomer heat transfer analysis between the corresponding battery module and the battery monomer in the three-dimensional physical field coupling model of the power battery is carried out based on the corresponding thermal temperature distribution of each monomer in the battery module and in combination with the heat flow distribution coefficient of the contact interface between the battery monomer and the module structure, so as to obtain the heat transfer path between each monomer in the battery module.

[0113] In the embodiment of the present application, by means of the thermal temperature distribution corresponding to each monomer in the battery module, combined with the heat flow distribution coefficient corresponding to the contact interface between the battery monomer and the module structure, the module monomer heat transfer analysis between the battery module and the battery monomer is carried out. Taking the charging time of 30 minutes as an example, the center temperature of monomer 1 is 45℃, the center temperature of monomer 2 is 43℃, the heat flow distribution coefficient of the contact interface between monomer 1 and the heat sink is 0.6, and the heat flow distribution coefficient of the contact interface between monomer 1 and the metal frame is 0.4. According to Fourier's law The total heat flow density of monomer 1 is calculated, assuming that the thermal conductivity k=2 W / (m K), temperature gradient After the total heat flow Q is calculated according to the temperature difference and distance between adjacent monomers, according to the heat flow distribution coefficient, the heat flow of 0.6Q is distributed to the contact interface with the heat sink, and the heat flow of 0.4Q is distributed to the contact interface with the metal frame. For the heat transfer between monomer 1 and monomer 2, the heat flow distribution coefficient of the contact interface between the two (assuming 0.5) is also considered, and the heat flow size and direction are calculated by combining the temperature difference. By calculating the contact interface between all monomers and structure in the module and the contact interface between monomers, the size and direction of heat flow at each place are determined, and by using three-dimensional visualization software, different colors and thickness of arrows are marked on the three-dimensional model of the battery module to show the heat transfer path. It can be clearly seen that heat is not only transferred between monomers, but also transferred to module structures through different contact interfaces in different proportions, such as from the middle monomer with higher temperature, part of the heat is quickly discharged through the heat sink, and part of the heat is transferred to the edge monomer through the metal frame. Finally, a complete and accurate heat transfer path between each monomer in the battery module is obtained, which provides detailed basis for optimizing the battery module thermal management scheme.

[0114] Further, step S3 comprises the following steps:

[0115] Step S31: based on the neural network combined with the dynamic battery thermal characteristic data set, a multi-scale deep neural network architecture is constructed, which includes an input layer, a hidden layer and an output layer;

[0116] In the embodiment of the present application, by building a multi-scale deep neural network architecture based on the constructed power battery thermal characteristic data set, the number of input layer neurons is determined according to the feature dimension of the data set. Since the power battery thermal characteristic data set integrates micro material thermal physical property data, battery monomer thermal characteristic data, battery module thermal distribution data and battery system thermal management parameter data, a total of 50 features such as electrode material thermal conductivity, electrolyte specific heat capacity, monomer temperature change curve data points, module monomer temperature difference, etc. are included. Therefore, 50 neurons are set in the input layer, each neuron corresponds to an input interface of a feature data, responsible for receiving and transmitting data. The hidden layer adopts a multi-layer structure, and 3 hidden layers are set, with 100, 80 and 60 neurons in each layer respectively. The neurons in each hidden layer are connected to the neurons in the previous layer through weights, and the layers are connected in a full connection manner. This structure design aims to mine the complex correlation between different scale thermal characteristic data through multi-layer nonlinear transformation, such as capturing the influence of micro material thermal physical property on battery monomer thermal characteristic, and the relationship between the performance of battery monomer thermal characteristic on the module scale and the system thermal management parameters. The output layer is set with 3 neurons, respectively corresponding to the output of three key parameters of power battery thermal management prediction: battery temperature distribution, heat generation efficiency and heat transfer coefficient. The output value of each neuron is processed by an activation function and output as part of the final prediction result, thereby constructing a complete multi-scale deep neural network architecture and laying a foundation for subsequent thermal management prediction training.

[0117] Step S32: thermal management prediction training of the multi-scale deep neural network architecture based on the power battery thermal characteristic data set, taking the power battery thermal characteristic data set as the training set to input the corresponding input layer of the multi-scale deep neural network architecture, and automatically assigning weights of different power battery thermal characteristics corresponding features in the hidden layer, while designing a hybrid loss to adjust the model parameters with an adaptive learning rate, wherein the hybrid loss includes temperature prediction error, heat generation rate prediction error and heat conduction coefficient prediction error, and the output layer simultaneously predicts the thermal management prediction parameters of the corresponding battery temperature distribution, heat generation efficiency and heat transfer coefficient of the power battery to train and generate a power battery thermal management prediction model;

[0118] In the embodiment of the present application, the constructed multi-scale deep neural network architecture is trained for thermal management prediction by using the power battery thermal characteristic dataset. The dataset is divided into a training set and a validation set in a ratio of 8:2, wherein the training set is used for model parameter adjustment, and the validation set is used for evaluating the model training effect. In the training process, the training set data is sequentially input into the input layer of the neural network. After the data enters the hidden layer, each neuron performs weighted summation on the input data according to the connection weight, and performs nonlinear transformation through an activation function (such as a ReLU function). The weights of different power battery thermal characteristics corresponding to the features are automatically adjusted in the hidden layer through a back propagation algorithm. For example, if it is found that the thermal conductivity coefficient of the electrode material has less influence on the battery temperature distribution prediction result at the beginning of the training, the algorithm will automatically reduce the corresponding weight. If the single cell temperature change curve data is crucial to the thermal generation efficiency prediction, the weight of this feature will be increased to highlight its role in the model. A hybrid loss function is designed to comprehensively consider the temperature prediction error, thermal generation rate prediction error and thermal conductivity coefficient prediction error. The mean square error (MSE) is used to calculate the error between the predicted values and the actual values of these three parameters, and then they are weighted and summed according to a certain proportion (such as temperature prediction error accounting for 40%, thermal generation rate prediction error accounting for 30%, and thermal conductivity coefficient prediction error accounting for 30%) to obtain the final hybrid loss value. An adaptive learning rate adjustment algorithm (such as the Adam algorithm) is used to dynamically adjust the learning rate according to the change of the hybrid loss value. In the early stage of training, a larger learning rate can speed up the update of model parameters. As the training progresses, when the loss value decreases to a certain extent, the learning rate is automatically reduced to avoid over updating of model parameters, ensuring the stability and accuracy of model training. After multiple rounds of training (the training period is set to 100 times), the prediction error of the model on the validation set gradually decreases and tends to be stable. Finally, the output layer outputs the predicted power battery thermal management parameters, including battery temperature distribution, thermal generation efficiency and thermal transfer coefficient, thereby training the power battery thermal management prediction model.

[0119] Step S33: Obtain the battery thermal characteristic dataset corresponding to the real-time power battery, and input the battery thermal characteristic dataset corresponding to the real-time power battery into the power battery thermal management prediction model for thermal management prediction to output the thermal management prediction result corresponding to the power battery.

[0120] In the embodiment of the present application, by acquiring the real-time battery thermal characteristic data set corresponding to the power battery in the actual application scene, real-time data is collected by various sensors installed in the battery system, such as temperature sensors collecting temperature data at different positions of the battery monomer every 10 seconds, pressure sensors monitoring the pressure change of the cooling medium to calculate the flow, and combining the charge and discharge current, voltage and other data recorded by the battery management system, after data preprocessing (including data cleaning, normalization and other operations), a real-time battery thermal characteristic data set consistent with the training set format is constructed, the real-time battery thermal characteristic data set is input into the trained power battery thermal management prediction model, and the data is processed in turn through the input layer and the hidden layer. In the hidden layer, the model extracts and analyzes the features of the input data according to the weights obtained by training, and mines the thermal characteristic information contained in the real-time data. Finally, the output layer outputs the corresponding thermal management prediction results of the power battery, including the predicted battery temperature distribution (such as predicting that the center temperature of the battery monomer will rise from 45℃ to 48℃ in the next 10 minutes), the heat generation efficiency (such as predicting that the heat generation efficiency under the current working condition is 500J per minute), and the heat transfer coefficient (such as predicting that the heat transfer coefficient between the battery module and the heat sink is 15 W / (m 2 K) ), These prediction results can provide decision basis for the power battery thermal management system, for example, according to the predicted temperature rise, the cooling medium flow is adjusted in advance, the thermal management strategy is optimized, and the power battery is ensured to run in a safe and efficient temperature range.

[0121] Further, step S31 includes the following steps:

[0122] The correlation between each power battery thermal characteristic parameter is obtained through the power battery thermal characteristic data set, and the number of nodes and the feature combination mode corresponding to the input layer are determined based on the correlation between each power battery thermal characteristic parameter. At the same time, a hierarchical feature extraction module is designed based on the neural network, so as to adopt different feature extraction methods for different power battery thermal characteristic parameters, including convolutional neural network CNN for processing spatial features and recurrent neural network RNN for processing time series features;

[0123] In the embodiment of the present application, by basing on the thermal characteristic data set of the power battery, first, the correlation analysis between parameters is carried out, the Pearson correlation coefficient calculation method is adopted, the 50 thermal characteristic parameters (such as the thermal conductivity of electrode material, the specific heat capacity of electrolyte, the data points of single cell temperature change curve, the temperature difference between single cells in module, etc.) in the data set are calculated two by two, and the correlation coefficient threshold is set to 0.7. When the absolute value of the correlation coefficient between two parameters is greater than the threshold, it is considered that the two parameters have strong correlation. Through calculation, it is found that the correlation between the thermal conductivity of electrode material and the temperature change rate of battery single cell in the charging and discharging process is 0.82, which belongs to strong correlation. While the correlation between the specific heat capacity of electrolyte and the heat generation efficiency is only 0.21, which is weak correlation. According to these correlation results, the number of input layer nodes and the feature combination method are determined. For the strongly correlated parameters, feature combination is carried out, such as combining the thermal conductivity of electrode material and the temperature change rate into a new feature dimension to reduce the number of input layer nodes. Finally, 40 nodes are determined for the input layer. Based on the neural network design, a hierarchical feature extraction module is designed. Different feature extraction methods are adopted for different types of power battery thermal characteristic parameters. For parameters with spatial characteristics, such as the temperature distribution data of each single cell in the battery module, convolutional neural network (CNN) is used for processing. The temperature distribution data is organized into a two-dimensional matrix form, a 3x3 convolution kernel is designed, and spatial features are extracted through convolution operation to capture the local pattern and spatial correlation of temperature distribution. For parameters with time sequence characteristics, such as the temperature change curve data of battery single cell in the charging and discharging process, recurrent neural network (RNN) is used for processing. The temperature change curve is segmented into sequence data in time sequence and input into RNN. Through the transmission mechanism of hidden state, the time sequence dependence and change trend in the data are captured.

[0124] Preferably, the hidden layer is designed and the residual block and attention mechanism are introduced therein, wherein the residual block is used to solve the gradient disappearance of deep network, and the attention mechanism is used to automatically assign the weight of different features;

[0125] In the embodiments of the present application, the residual block is introduced when designing the hidden layer to solve the gradient vanishing problem of deep network. The hidden layer is divided into multiple residual units, each of which contains two fully connected layers and a skip connection. Taking the first residual unit as an example, the input data first passes through the first fully connected layer, which contains 100 neurons and uses the ReLU activation function for nonlinear transformation. The transformed output then passes through the second fully connected layer, which also contains 100 neurons and the ReLU activation function. Then, the original input data is directly added to the output of the second fully connected layer through the skip connection to form the residual connection. This structure allows the gradient to be directly transmitted through the skip connection during the training process, avoiding the problem of gradient vanishing in deep network, and improving the training efficiency and stability of the network. At the same time, an attention mechanism is introduced in the hidden layer to automatically assign weights to different features. After the output of each residual unit, an attention module is added. The module first performs linear transformation on the input features, and then calculates the attention weight of each feature through the softmax function. For example, for the battery temperature distribution prediction task, if the temperature change curve data at a certain time is more important for the prediction result, the attention mechanism will automatically assign a higher weight to this data. The specific calculation process is as follows: the output of the residual unit is input into a fully connected layer containing 50 neurons, processed by the tanh activation function, and then input into another fully connected layer containing 40 neurons. Finally, the attention weight of each feature is obtained through the softmax function. Multiply these weights with the original features to get the attention-adjusted feature representation, thereby highlighting the role of important features and improving the prediction accuracy of the model.

[0126] Preferably, the corresponding output layer is designed for multi-task learning to simultaneously predict the thermal management prediction parameters of the corresponding battery temperature distribution, heat generation efficiency and heat transfer coefficient of the power battery, and the corresponding multi-scale deep neural network architecture is constructed by the input layer, the hidden layer and the output layer.

[0127] In the embodiment of the present application, by designing a multi-task learning corresponding output layer to simultaneously predict the battery temperature distribution, heat generation efficiency and heat transfer coefficient of the power battery, the output layer is composed of three independent sub-layers, each of which is responsible for predicting a parameter. For the battery temperature distribution prediction sub-layer, the output of the hidden layer is first transformed by a fully connected layer containing 80 neurons, then by a fully connected layer containing 40 neurons, and finally connected to an output layer containing 10 neurons, each neuron corresponding to a temperature prediction value at different positions in the battery module. The mean square error loss function is used to calculate the error between the predicted temperature and the actual temperature. For the heat generation efficiency prediction sub-layer, the output of the hidden layer is transformed by a fully connected layer containing 60 neurons, and then connected to a fully connected layer containing 30 neurons, and finally output the heat generation efficiency prediction value through a neuron. The mean absolute error loss function is used to evaluate the accuracy of the prediction result. For the heat transfer coefficient prediction sub-layer, the output of the hidden layer is sequentially transformed by fully connected layers containing 70 and 35 neurons, and finally output the heat transfer coefficient prediction value through a neuron. Similarly, the mean absolute error loss function is used. The complete multi-scale deep neural network architecture is constructed by the input layer (40 nodes), the hidden layer (containing multiple residual units and attention modules) and the output layer (three independent sub-layers). This architecture can fully utilize different types of feature extraction methods to mine the complex relationships in the power battery thermal characteristic data, and simultaneously predict multiple thermal management parameters, providing a strong model foundation for subsequent thermal management prediction training.

[0128] Further, step S4 comprises the following steps:

[0129] Step S41: based on the thermal management prediction result corresponding to the power battery, the thermal management state change of the thermal management system corresponding to the power battery is analyzed to obtain the power battery thermal management state change space data;

[0130] In the embodiment of the present application, by analyzing the state change of the thermal management system based on the thermal management prediction result of the power battery, taking the prediction result at a certain time as an example, it is predicted that the battery module center temperature will rise from 45℃ to 48℃, the edge monomer temperature will rise from 42℃ to 44℃, the heat generation efficiency will be 500J per minute, and the heat transfer coefficient will be 15 W / (m 2 K), the predicted data is mapped into the three-dimensional model of the thermal management system of the power battery by three-dimensional visualization technology to generate thermal management state change space data, in which different colors represent different temperature regions and arrows represent heat flow directions, and color depth and arrow thickness reflect temperature values and heat flow sizes. For example, a high-temperature region is represented by a red region, a low-temperature region is represented by a blue region, and an arrow pointing from a high-temperature central region to a low-temperature edge region represents a heat transfer direction. Further analysis of the temperature distribution trend calculates the temperature gradient, which is (48℃-44℃) / 0.1m=40K / m (assuming the size of the module is 0.1m) under the current predicted condition. At the same time, the influence of the change of heat generation efficiency and heat transfer coefficient on the overall thermal state is analyzed, and the increase of the heat generation efficiency will cause the temperature rise speed to accelerate, and the size of the heat transfer coefficient directly affects the heat dissipation efficiency. In this way, the thermal management state change space data of the power battery is comprehensively obtained, which provides a basis for the subsequent definition of the thermal management objective function.

[0131] Step S42: defining a thermal management objective function corresponding to the power battery according to the thermal management state change space data of the power battery, wherein the thermal management objective function includes optimization objective functions corresponding to temperature distribution uniformity, energy consumption and power battery life;

[0132] In the embodiment of the present application, by defining the thermal management objective function according to the thermal management state change space data of the power battery, firstly, for the temperature distribution uniformity, the objective function is constructed, assuming that the battery module contains monomers, the temperature of the th monomer is , and the average value of the monomer temperature is , the standard deviation of the temperature of each monomer in the battery module is calculated, which is used as an evaluation index of the temperature distribution uniformity, and the objective is to minimize the standard deviation, i.e. the temperature distribution is as uniform as possible. For example, under the current predicted state, the standard deviation of the temperature of 12 monomers is 1.8℃, and the objective is to reduce it to below 1.0℃ through thermal management control. For the energy consumption objective function, the power consumption of the cooling pump is proportional to the cube of the flow rate ( is a proportional coefficient), the power consumption of the fan is proportional to the cube of the rotation speed ( is a proportional coefficient), and the power consumption of the heating element is proportional to the square of the current ( by establishing the mathematical relationship between the power consumption of these devices and the control parameters (such as cooling pump speed, fan power, heating element current), an energy consumption objective function is constructed The target is to minimize the total energy consumption under the premise of meeting the temperature control requirements. For the power battery life objective function, the impact of temperature on battery life is considered. Studies have shown that working at too high or too low temperature will accelerate aging and shorten the life of the battery. Through experimental data fitting, a relationship model between temperature and battery cycle life is established. For example, when the average temperature of the battery is between 25℃-35℃, the cycle life is the longest, and the life will gradually shorten when deviating from this range. The target is to keep the battery working temperature in this optimal range as much as possible, that is Wherein is the battery life, , is the fitting coefficient, is the average temperature, is the optimal working temperature. To maximize the battery life, the three aspects are integrated to construct a complete thermal management objective function , , and are weight coefficients.

[0133] Step S43: The thermal management system corresponding to the power battery is designed to include thermal management constraint conditions corresponding to temperature upper and lower limit constraints, cooling system power constraints and charge and discharge current constraints, and the thermal management constraint optimization control is performed on the thermal management objective function corresponding to the power battery based on the thermal management constraint conditions, to generate an optimal thermal management control strategy of the power battery;

[0134] In the embodiment of the present application, by designing the thermal management constraint condition, the thermal management objective function is constrained and optimized. First, the upper and lower temperature constraints are set. According to the specification requirements of the battery manufacturer, the upper limit of the temperature of the battery monomer is 55℃, and the lower limit is 15℃. In the optimization process, the temperature of all monomers is ensured to be within this range. The cooling system power constraint is set. The maximum power of the cooling pump is 500W, the maximum power of the fan is 300W, and the maximum power of the heating element is 200W. In the optimization process, the total power consumption of the cooling system cannot exceed these limits. The charging and discharging current constraint is set. According to the performance characteristics of the battery, the maximum charging current is 2C (40A), and the maximum discharging current is 3C (60A). In the thermal management process, the charging and discharging current must meet these constraint conditions. Based on these constraint conditions, the thermal management objective function is optimized and solved by using the sequential quadratic programming algorithm. The algorithm gradually finds the optimal solution that meets all the constraint conditions through iteration. In each iteration, the gradient and Hessian matrix of the objective function are calculated, and the value of the control parameter is updated according to this information. After multiple iterations, the optimal thermal management control parameters are obtained, including the optimal speed of the cooling pump, the optimal power of the fan and the optimal current of the heating element, and the optimal thermal management control strategy of the power battery is generated.

[0135] Step S44: uploading the optimal thermal management control strategy of the power battery to the cloud platform and converting it into actual thermal management system control parameters, including the cooling pump speed, fan power and heating element current, responding to the control of the corresponding thermal management system of the power battery to perform corresponding thermal management control work.

[0136] In the embodiment of the present application, the generated optimal thermal management control strategy of the power battery is uploaded to the cloud platform to convert the optimized control parameters (such as the cooling pump speed of 1200 rpm, the fan power of 150 W, and the heating element current of 2 A) into a standard data format through a dedicated data transmission protocol, and send them to the server of the cloud platform. After receiving the control parameters, the cloud platform performs format verification and parameter rationality check to verify whether the cooling pump speed is within the allowed operating range of the device (such as 500-3000 rpm), whether the fan power exceeds the maximum power limit (300 W), and whether the heating element current meets the safety standard (such as not exceeding 5 A). After verification, the cloud platform converts the control parameters into actual thermal management system control signals. For the cooling pump, the speed parameter is converted into a corresponding voltage control signal to adjust the supply voltage of the cooling pump through the frequency converter to realize speed control. For the fan, the power parameter is converted into a PWM (Pulse Width Modulation) control signal to control the speed of the fan by adjusting the duty cycle of the PWM signal, thereby realizing power regulation. For the heating element, the current parameter is converted into a current control signal to control the current size of the heating element by adjusting the resistance in the circuit or using a constant current source. After receiving these control signals, the thermal management system performs the corresponding thermal management control work. The cooling pump operates at a speed of 1200 rpm to provide stable cooling medium flow; the fan works at a power of 150 W to accelerate air flow and enhance heat dissipation effect; the heating element works at a current of 2 A to provide appropriate heating for the battery in a low temperature environment. In this way, precise control of the power battery thermal management system is realized to ensure that the battery works in a safe and efficient temperature range.

[0137] Further, the present application also provides a neural network-based power battery thermal management simulation system for executing the neural network-based power battery thermal management simulation method as described above, which comprises:

[0138] a three-dimensional coupling simulation construction module for acquiring corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery through coupling experimental measurement, and constructing a three-dimensional physical field coupling model of the power battery based on the corresponding electrical performance parameters, thermal performance parameters and mechanical performance parameters of the power battery, thereby generating the three-dimensional physical field coupling model of the power battery;

[0139] a battery thermal characteristic evaluation module for evaluating the battery thermal characteristics from the micro material, the battery monomer, the battery module and the thermal management system of the power battery according to the three-dimensional physical field coupling model of the power battery to obtain a battery thermal characteristic data set, which includes micro material thermal physical property data, battery monomer thermal characteristic data, battery module thermal distribution data and battery system thermal management parameter data;

[0140] a thermal management prediction module, configured to construct a multi-scale deep neural network architecture based on a neural network, and perform thermal management prediction training on the multi-scale deep neural network architecture based on a power battery thermal characteristic dataset, to generate a power battery thermal management prediction model, and output a corresponding thermal management prediction result of the power battery, which includes thermal management prediction parameters corresponding to a battery temperature distribution, a heat generation efficiency, and a heat transfer coefficient;

[0141] a thermal management constraint control module, configured to perform thermal management constraint optimization control on a thermal management system corresponding to the power battery based on the corresponding thermal management prediction result of the power battery, to generate an optimal thermal management control strategy of the power battery, and upload the optimal thermal management control strategy of the power battery to a cloud platform to control the thermal management system corresponding to the power battery to perform corresponding thermal management control work.

[0142] The above description is merely that of the specific embodiments of the present application, which enables those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the application. Accordingly, the present application is not to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A simulation method for thermal management of power batteries based on neural networks, characterized in that, Includes the following steps: Step S1: Obtain the electrical, thermal, and mechanical performance parameters of the power battery through coupling experiments, and construct a three-dimensional coupling simulation of the power battery based on the electrical, thermal, and mechanical performance parameters of the power battery to generate a three-dimensional physical field coupling model of the power battery. Step S2: Using a three-dimensional physical field coupling model of the power battery, the thermal characteristics of the battery are evaluated at four scales: micromaterials, individual cells, battery modules, and thermal management systems, to obtain a power battery thermal characteristic dataset. This dataset includes micromaterial thermal property data, individual cell thermal characteristic data, battery module thermal distribution data, and battery system thermal management parameter data. Step S2 includes the following steps: Step S21: Extract battery thermal parameters from the microscopic material scale corresponding to the power battery using the three-dimensional physical field coupling model of the power battery to obtain microscopic material thermal property data, including the thermal conductivity of electrode materials and the specific heat capacity of electrolyte. Step S22: Using the three-dimensional physical field coupling model of the power battery, perform single-cell charge-discharge thermal analysis at the scale of the corresponding battery cell to obtain the thermal characteristic data of the battery cell, including the temperature change curve and internal resistance heat generation rate during the charge-discharge process. Step S23: Using the three-dimensional physical field coupling model of the power battery, perform module thermal distribution analysis from the scale of the corresponding battery module to obtain battery module thermal distribution data, including the temperature difference between individual cells in the battery module and the heat transfer path. Step S24: Perform system thermal management statistical analysis from the scale of the corresponding thermal management system of the power battery using the three-dimensional physical field coupling model of the power battery to obtain battery system thermal management parameter data, including cooling medium flow rate and inlet temperature distribution. Step S25: Merge the thermal property data of micromaterials, the thermal characteristic data of individual cells, the thermal distribution data of battery modules, and the thermal management parameter data of battery system into the same dataset to obtain the thermal characteristic dataset of power battery. Step S3: Construct a multi-scale deep neural network architecture based on the power battery thermal characteristic dataset, and train the multi-scale deep neural network architecture for thermal management prediction based on the power battery thermal characteristic dataset to generate a power battery thermal management prediction model and output the corresponding thermal management prediction results of the power battery, including the thermal management prediction parameters corresponding to the battery temperature distribution, heat generation efficiency and heat transfer coefficient. Step S4: Based on the thermal management prediction results corresponding to the power battery, perform thermal management constraint optimization control on the thermal management system corresponding to the power battery to generate the optimal thermal management control strategy for the power battery; upload the optimal thermal management control strategy for the power battery to the cloud platform to respond to control the thermal management system corresponding to the power battery to execute the corresponding thermal management control work.

2. The neural network-based power battery thermal management simulation method according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain the electrical, thermal, and mechanical performance parameters of the power battery through a coupling experiment. Step S12: Obtain the design structure data corresponding to each component of the power battery, and construct the topology connection of the power battery based on the design structure data corresponding to each component of the power battery to generate the topology connection architecture of each component of the power battery. Step S13: Perform physical field simulation analysis on the power battery based on the electrical performance parameters, thermal performance parameters and mechanical performance parameters corresponding to the power battery, and generate the electrical simulation field, thermal simulation field and mechanical simulation field corresponding to the power battery. Step S14: Based on the electrical simulation field, thermal simulation field and mechanical simulation field corresponding to the power battery, perform three-dimensional mapping and coupling construction of the topological connection architecture of each component of the power battery to generate a three-dimensional physical field coupling model of the power battery.

3. The neural network-based power battery thermal management simulation method according to claim 1, characterized in that, Step S22 includes the following steps: By constructing the cell material tensor at the scale of the corresponding cell using a three-dimensional physical field coupling model of the power battery, the microstructure changes of the electrode and electrolyte materials on the cell under different temperatures and charging states are analyzed. The conduction paths of ions and electrons in the material are abstracted into tensor form to construct the electrode-electrolyte microstructure conductivity tensor, where the tensor elements represent the conductivity values ​​oriented inside the corresponding material. Based on the conductivity tensor of the electrode-electrolyte microstructure and combined with the entropy change theory in thermodynamics, an ion migration network is generated for the corresponding battery cell in the three-dimensional physical field coupling model of the power battery. This study investigates the migration behavior of ions between the electrode and electrolyte during the charging and discharging process of the battery cell and regards this migration behavior as flux transport driven by entropy change. The interface between the electrode and electrolyte and the internal pores are used as nodes, and the ion migration path is used as the edge. By calculating the entropy change value of each node and the flux of the edge, the corresponding ion migration entropy change flux network is constructed. By considering the contact characteristics between the current collector and the electrode material in the corresponding battery cell within the three-dimensional physical field coupling model of the power battery, and introducing the concept of contact thermal resistance in heat transfer, the surface roughness and contact pressure distribution of the current collector and the electrode material are analyzed. Based on the surface roughness and contact pressure distribution of the current collector and the electrode material, the thermal resistance of each contact area between the current collector and the electrode material is calculated to construct the current collector-electrode contact thermal resistance matrix. Based on the ion migration entropy change flux network and the current collector-electrode contact thermal resistance matrix, the charging and discharging heat source analysis of the corresponding battery cell in the three-dimensional physical field coupling model of the power battery is carried out. The electrochemical reaction heat of the battery cell during the charging and discharging process is calculated based on the entropy change and reaction enthalpy change during the ion migration process. The Joule heat of the battery cell during the charging and discharging process is solved by the conductivity tensor, current distribution and thermal resistance of the electrode-electrolyte microstructure. The electrochemical reaction heat and Joule heat are coupled in the three-dimensional space of the battery cell to construct the electrochemical reaction heat-Joule heat coupled heat source. Based on the electrochemical reaction heat-Joule heat coupling heat source and combined with the geometry and microscopic material thermal property data of the battery cell, an unsteady-state heat transfer analysis is performed on the corresponding battery cell in the three-dimensional physical field coupling model of the power battery. The thermal conductivity temperature distribution is solved by the finite difference method or finite element method to obtain the temperature distribution of the battery cell at different times and locations during the charging and discharging process. The temperature distribution of the battery cell at different times and locations during the charging and discharging process is fitted into an unsteady-state heat transfer three-dimensional surface. The points on this surface correspond to the spatial location and temperature distribution of the battery cell. The change trend of the surface forms a temperature change curve. At the same time, the internal resistance heating rate is obtained by calculating the heat flux density gradient.

4. The neural network-based power battery thermal management simulation method according to claim 1, characterized in that, Step S23 includes the following steps: By using a three-dimensional physical field coupling model of the power battery, the thermal temperature distribution of each cell in the module is analyzed at the scale of the corresponding battery module to obtain the thermal temperature distribution of each cell in the battery module. The temperature difference is quantified based on the thermal temperature distribution of each cell in the battery module to obtain the temperature difference between each cell in the battery module. Based on the thermal temperature distribution of each cell within the battery module, the heat transfer between the battery module and individual cells in the three-dimensional physical field coupling model of the power battery is analyzed to obtain the heat transfer path between each cell within the battery module.

5. The neural network-based power battery thermal management simulation method according to claim 4, characterized in that, The analysis of heat transfer between the battery module and individual battery cells in the three-dimensional physical field coupling model of the power battery based on the thermal temperature distribution of each cell within the battery module includes the following steps: The contact interface between the battery cell and the module structure is determined by the corresponding battery module and battery cell in the three-dimensional physical field coupling model of the power battery. The corresponding material surface roughness measurement data and contact pressure distribution are obtained by the corresponding battery module in the three-dimensional physical field coupling model of the power battery. Based on the material surface roughness measurement data and contact pressure distribution, the heat flow distribution of the contact interface between the battery cell and the module structure is evaluated to obtain the heat flow distribution coefficient corresponding to the contact interface between the battery cell and the module structure, where the heat flow ratio from the battery cell to the module structure at the contact interface is represented. Based on the thermal temperature distribution of each cell within the battery module and the heat flow distribution coefficient at the contact interface between the battery cell and the module structure, the heat transfer between the battery module and the battery cells in the three-dimensional physical field coupling model of the power battery is analyzed to obtain the heat transfer path between each cell within the battery module.

6. The neural network-based power battery thermal management simulation method according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Construct a multi-scale deep neural network architecture based on the neural network and the power battery thermal characteristic dataset, including an input layer, hidden layers and an output layer; Step S32: Based on the power battery thermal characteristic dataset, a multi-scale deep neural network architecture is trained for thermal management prediction. The power battery thermal characteristic dataset is used as the training set to input the input layer of the multi-scale deep neural network architecture for training. The weights of the features corresponding to different power battery thermal characteristics are automatically assigned in the hidden layer. At the same time, a hybrid loss is designed and an adaptive learning rate is used to adjust the model parameters. The hybrid loss includes temperature prediction error, heat generation rate prediction error and heat transfer coefficient prediction error. The output layer also predicts the thermal management prediction parameters of the power battery corresponding to the battery temperature distribution, heat generation efficiency and heat transfer coefficient to train and generate a power battery thermal management prediction model. Step S33: Obtain the real-time battery thermal characteristic dataset corresponding to the power battery, and input the real-time battery thermal characteristic dataset corresponding to the power battery into the power battery thermal management prediction model to perform thermal management prediction, so as to output the thermal management prediction result corresponding to the power battery.

7. The neural network-based power battery thermal management simulation method according to claim 6, characterized in that, Step S31 includes the following steps: The correlation between various thermal characteristic parameters of power batteries is obtained by using a power battery thermal characteristic dataset. Based on the correlation between various thermal characteristic parameters of power batteries, the number of nodes and feature combination methods of the input layer are determined. At the same time, a hierarchical feature extraction module is designed based on a neural network to use different feature extraction methods for different thermal characteristic parameters of power batteries, including convolutional neural networks (CNN) for processing spatial features and recurrent neural networks (RNN) for processing temporal features. By designing hidden layers and introducing residual blocks and attention mechanisms, the residual blocks are used to solve the gradient vanishing problem in deep networks, while the attention mechanism is used to automatically assign weights to different features. By designing a multi-task learning output layer, thermal management prediction parameters such as battery temperature distribution, heat generation efficiency, and heat transfer coefficient of the power battery are predicted simultaneously. A corresponding multi-scale deep neural network architecture is constructed through the input layer, hidden layer, and output layer.

8. The neural network-based power battery thermal management simulation method according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Based on the thermal management prediction results corresponding to the power battery, perform thermal management state change analysis on the thermal management system corresponding to the power battery to obtain the spatial data of thermal management state change of the power battery. Step S42: Define the thermal management objective function corresponding to the power battery based on the spatial data of the thermal management state change of the power battery, including the optimization objective functions corresponding to uniform temperature distribution, energy consumption and power battery life; Step S43: Design the thermal management system corresponding to the power battery, including the upper and lower temperature limits, cooling system power constraints, and charge and discharge current constraints, and perform thermal management constraint optimization control on the thermal management objective function corresponding to the power battery based on the thermal management constraints to generate the optimal thermal management control strategy for the power battery. Step S44: Upload the optimal thermal management control strategy of the power battery to the cloud platform and convert it into actual thermal management system control parameters, including cooling pump speed, fan power and heating element current, and respond to control the corresponding thermal management system of the power battery to perform the corresponding thermal management control work.

9. A power battery thermal management simulation system based on neural networks, characterized in that, For executing the neural network-based power battery thermal management simulation method as described in claim 1, the neural network-based power battery thermal management simulation system includes: The three-dimensional coupled simulation construction module is used to obtain the electrical, thermal and mechanical performance parameters of the power battery through coupled experimental measurement, and to construct a three-dimensional coupled simulation of the power battery based on the electrical, thermal and mechanical performance parameters of the power battery, thereby generating a three-dimensional physical field coupled model of the power battery. The battery thermal characteristic evaluation module is used to evaluate the battery thermal characteristics at four scales: micromaterials, battery cells, battery modules, and thermal management systems, using a three-dimensional physical field coupling model of the power battery. This results in a power battery thermal characteristic dataset, which includes micromaterial thermal property data, battery cell thermal characteristic data, battery module thermal distribution data, and battery system thermal management parameter data. The thermal management prediction module is used to build a multi-scale deep neural network architecture based on a neural network, and to train the multi-scale deep neural network architecture for thermal management prediction based on a power battery thermal characteristic dataset, so as to generate a power battery thermal management prediction model and output the corresponding thermal management prediction results of the power battery, including the thermal management prediction parameters corresponding to the battery temperature distribution, heat generation efficiency and heat transfer coefficient. The thermal management constraint control module is used to optimize the thermal management system of the power battery based on the thermal management prediction results of the power battery, so as to generate the optimal thermal management control strategy of the power battery; and upload the optimal thermal management control strategy of the power battery to the cloud platform to respond to control the thermal management system of the power battery to execute the corresponding thermal management control work.

Images