Battery module temperature distribution inversion, inversion model construction method and device

Abstract

This invention discloses a method and apparatus for inverting the temperature distribution of a battery module and constructing an inversion model. The method includes: establishing a geometric model based on the geometric characteristics of an experimental battery pack; establishing a finite element temperature simulation model of the experimental battery module including the geometric model and an electrothermal coupled physical field; simulating the operating state of the battery module based on the constructed simulation model; collecting surface and internal temperature data during the simulation process; training the inversion model of the power battery temperature based on the obtained surface and internal temperature data; and finally obtaining a trained temperature distribution inversion model of the experimental battery module, which is then applied to actual measurements. This method can better predict the internal temperature of the battery based on the battery surface temperature, obtain more accurate temperature prediction results, and more timely and accurate predict potential overheating states of the battery, thereby improving the safety of the power battery and ensuring the safety of electric vehicles.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle equipment, and in particular to a method and apparatus for inverting the temperature distribution of battery modules and constructing an inversion model. Background Technology

[0002] As the energy storage device and power source for electric vehicles, the performance and lifespan of the power battery pack are greatly affected by temperature. The safe and stable operation of the battery pack is crucial to the safety and reliability of the electric vehicle. However, potential faults such as overheating and mechanical puncture during storage and charging / discharging of the electric vehicle's power battery pack can lead to thermal runaway. Therefore, measuring the temperature of the electric vehicle's power battery pack and calculating its internal temperature distribution is essential for assessing the electric vehicle's operating status and preventing thermal runaway caused by battery overheating and mechanical puncture. This is of great significance for the operation and maintenance of electric vehicle power batteries.

[0003] In existing technologies, the prediction of battery overheating faults is usually based on the current operating data of the battery pack, using random forest regression technology to predict the probability of overheating. For example, Chinese patent application CN115958957, entitled "A Method and System for Predicting Overheating Faults of Electric Vehicle Power Batteries," discloses a method and system for predicting overheating faults of electric vehicle power batteries. This method collects data on the current operation of the power battery pack, trains a random forest big data regression model based on a preset historical dataset of the power battery's operating status under various charging conditions, predicts the battery's operating data within a preset future time, and uses the future operating data as a fault discrimination index to obtain the probability of the battery overheating fault occurring within the preset future time. Summary of the Invention

[0004] Electric vehicle battery packs may experience thermal runaway due to potential overheating, mechanical puncture, or other faults during storage and charging / discharging. Existing battery overheating fault prediction methods rely on purely data-driven approaches, which struggle to acquire massive amounts of applicable and effective data in real-world operations. Furthermore, they neglect the battery's operating principles and the physical relationships between its parameters, resulting in poor prediction accuracy and data utilization efficiency. Moreover, existing methods are unsuitable for scenarios with only battery modules and lacking a corresponding battery management system. In summary, existing battery pack overheating fault prediction methods cannot accurately predict battery overheating conditions, rely heavily on the battery management system, and have limited application scenarios and poor versatility.

[0005] In view of the above problems, the present invention is proposed to provide a method and apparatus for inverting the temperature distribution of a battery module to overcome or at least partially solve the above problems.

[0006] In a first aspect, embodiments of the present invention provide a method for constructing a battery module temperature distribution inversion model, comprising:

[0007] The temperature simulation results are obtained by using the finite element temperature simulation model of the experimental battery module to simulate the temperature under multiple loading conditions of the experimental battery module under preset working conditions. The finite element temperature simulation model includes the geometric model of the experimental battery module and the electrothermal coupling physical field constructed based on the geometric model. The electrothermal coupling physical field includes at least the electric field distribution relationship, the temperature distribution relationship, and the total accumulated heat determined based on the relationship.

[0008] Based on the temperature simulation results, a temperature distribution dataset was obtained, which includes each set of loading conditions and the corresponding temperature distribution data; the temperature distribution data includes the surface temperature and internal temperature of the battery module.

[0009] The temperature distribution dataset was used to train a pre-established inversion model of power battery temperature to obtain the temperature distribution inversion model of the experimental battery module.

[0010] In some optional embodiments, the finite element temperature simulation model of the experimental battery module is established as follows:

[0011] Based on the geometric features of the experimental battery module, a geometric model of the experimental battery module is constructed.

[0012] The grid data of the geometric model of the battery module is obtained by meshing the geometric model using a mesh partitioning method.

[0013] The Maxwell equations and solid heat transfer equations of the discretized experimental battery module are used to construct an electrothermal coupled physical field based on the grid data, the discretized equations, and the discretized solid heat transfer equations.

[0014] In some optional embodiments, a finite element temperature simulation model of the experimental battery module is used to simulate the temperature under multiple loading conditions of the experimental battery module under preset operating conditions, and the temperature simulation results are obtained, including:

[0015] Set the operating conditions of the experimental battery module and multiple sets of loading conditions under different operating conditions;

[0016] The finite element temperature simulation model of the experimental battery module was used to simulate the battery's operating state under different loading conditions; and the surface temperature and internal temperature of the experimental battery module under loading conditions were collected to obtain temperature distribution data of the experimental battery module under multiple sets of loading conditions under preset working conditions.

[0017] The preset operating conditions include at least one of normal operating conditions and fault operating conditions; the loading conditions include at least one of the following: ambient temperature of the simulation model, internal short-circuit fault conditions, and per-unit value of charging power.

[0018] In some optional embodiments, a temperature distribution dataset is obtained based on the temperature simulation results, including each set of loading conditions and the corresponding temperature distribution data; including:

[0019] Acquire temperature distribution data for the experimental battery modules under each loading condition; the temperature distribution data includes the internal temperature of the experimental battery modules and their corresponding surface temperatures;

[0020] Based on each set of loading conditions and the corresponding temperature distribution data, multiple sets of first temperature distribution arrays are established.

[0021] The experimental battery module temperature distribution dataset is constructed based on multiple sets of first temperature distribution arrays.

[0022] In some optional embodiments, the above method further includes:

[0023] The temperatures in each group of the first temperature distribution array in the temperature distribution dataset are normalized using the following formula: in For the normalized temperature data, T i Temperature data for the selected temperature points; T max With T min These are the maximum and minimum temperatures in the first temperature distribution array, respectively.

[0024] In some optional embodiments, a pre-established inversion model of the power battery temperature is trained using a temperature distribution dataset to obtain a temperature distribution inversion model of the experimental battery module, including:

[0025] According to the accuracy requirements, the temperature distribution dataset is divided into N1 groups of first temperature distribution subarrays under normal operating conditions and N2 groups of first temperature distribution subarrays under fault operating conditions. Each first temperature distribution subarray is represented by a feature matrix consisting of M data points.

[0026] The first temperature distribution subarray of group N1 under normal operating conditions and the first temperature distribution subarray of group N2 under fault operating conditions are respectively input into the inversion model of power battery temperature pre-built based on backpropagation neural network. The internal temperature prediction value of the experimental battery module is output. Based on the internal temperature prediction value and the internal temperature of the experimental battery module included in the temperature distribution dataset, the parameters of the inversion model are tuned, and the model training process is repeated until the model meets the preset convergence condition. The trained power battery temperature inversion model is obtained as the temperature distribution inversion model of the experimental battery module.

[0027] In some optional embodiments, the inversion model is parameter-tuned based on the predicted internal temperature and the internal temperature of the experimental battery module included in the temperature distribution subarray, and the model training process is repeated until the model meets the preset convergence conditions, including:

[0028] The error between the predicted internal temperature output by the inversion model and the measured internal temperature of the experimental battery module included in the temperature distribution dataset is calculated based on the pre-constructed error function. The partial derivative of the error with respect to the initial weights of the inversion model is also calculated to obtain the weight adjustment gradient of the inversion model. The weights are then updated based on the weight adjustment gradient. The error function of the inversion model is as follows: Among them O i t represents the predicted internal temperature output by the inversion model. i The measured internal temperature values ​​of the experimental battery module obtained from the simulation model;

[0029] The process of model training and error calculation is repeated multiple times until the error reaches the preset requirement, thus completing the training.

[0030] Secondly, embodiments of the present invention provide a method for inverting the temperature distribution of a battery module, comprising:

[0031] Collect surface temperature distribution data of the experimental battery module to be predicted under the preset loading conditions, input the preset loading conditions and the corresponding surface temperature distribution data into the trained temperature distribution inversion model of the experimental battery module, and output the internal temperature of the experimental battery module under the preset conditions.

[0032] The temperature distribution inversion model was trained using the aforementioned battery module temperature distribution inversion model construction method.

[0033] In some optional embodiments, surface temperature distribution data of the experimental battery module under preset loading conditions are collected, including:

[0034] Under preset operating conditions and preset loading conditions, the temperature at points on the surface of the battery module is obtained by temperature sensors arranged in the finite element temperature simulation model of the experimental battery module, thus obtaining the surface temperature distribution data of the experimental battery module under preset operating conditions and preset loading conditions.

[0035] In some optional embodiments, the temperature sensors arranged in the finite element temperature simulation model of the experimental battery module include:

[0036] A temperature sensor A is placed at the positive terminal of the battery, B temperature sensor B is placed at the negative terminal of the battery, and C temperature sensor C is placed on the side of the battery. The C temperature sensors placed on the side of the battery are evenly distributed.

[0037] Thirdly, embodiments of the present invention provide a battery module temperature distribution inversion model construction apparatus, comprising:

[0038] The data acquisition module is used to acquire temperature simulation results of the experimental battery module under multiple loading conditions based on a finite element temperature simulation model. The finite element temperature simulation model includes a geometric model of the experimental battery pack and an electrothermal coupling physical field constructed based on the geometric model. The electrothermal coupling physical field includes at least an electric field distribution relationship, a temperature distribution relationship, and a total accumulated heat determined based on the relationships. Based on the temperature simulation results, a temperature distribution dataset is obtained, including each loading condition and the corresponding temperature distribution data. The temperature distribution data includes the surface temperature and internal temperature of the battery module.

[0039] The model building module is used to train a pre-established inversion model of power battery temperature using a temperature distribution dataset to obtain a temperature distribution inversion model of the experimental battery module.

[0040] Fourthly, embodiments of the present invention provide a battery module temperature distribution inversion device, comprising:

[0041] The data acquisition module is used to collect surface temperature distribution data of the experimental battery module to be predicted under preset loading conditions.

[0042] The temperature prediction module is used to input the loading conditions of the preset working conditions and the corresponding surface temperature distribution data into the trained temperature distribution inversion model of the experimental battery module, and output the internal temperature of the experimental battery module under the preset working conditions; the temperature distribution inversion model is trained using the battery module temperature distribution inversion model construction method.

[0043] This invention provides a computer storage medium storing computer-executable instructions. When these computer-executable instructions are executed by a processor, they implement a method for constructing a battery module temperature distribution inversion model or a method for inverting battery module temperature distribution.

[0044] This invention provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements a method for constructing a battery module temperature distribution inversion model or a method for inverting battery module temperature distribution.

[0045] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0046] The battery module temperature distribution inversion model construction method provided in this invention establishes a geometric model based on the geometric characteristics of the experimental battery pack, and constructs an electrothermal coupled physical field based on the geometric model to reflect the physical laws of operation and battery operating state of the experimental battery pack. Finally, a finite element temperature simulation model of the experimental battery module including the geometric model and the electrothermal coupled physical field is established. The operating state of the battery module is simulated based on the constructed simulation model, and surface and internal temperature data are collected during the simulation process. The temperature data obtained in this way can reflect the physical relationship between various parameters of the battery and is closer to the real situation of the experimental battery pack. The inversion model of power battery temperature is trained based on the obtained surface and internal temperature data. The trained model can better predict the internal temperature of the battery based on the surface temperature of the battery and obtain more accurate temperature prediction results, and more timely and accurate prediction of possible overheating states of the battery.

[0047] The battery module temperature distribution inversion method provided in this embodiment of the invention predicts the internal temperature of the experimental battery pack based on the model trained by the above training method and the surface temperature distribution data of the experimental battery pack under the preset working conditions. This allows for accurate temperature prediction results, timely prediction of internal temperature changes in the battery, and timely warning and corresponding handling when overheating may occur. This helps to avoid accidents caused by overheating, improve the safety of the power battery, and ultimately ensure the safety of electric vehicles.

[0048] The aforementioned method for constructing and retrieving battery module temperature distribution inversion models, compared to traditional methods, incorporates the characteristics of a physical model-driven approach. This enables the generation of massive amounts of data based on physical models, avoiding the limitations of limited or incomplete battery operating status data in traditional methods, thus resulting in more accurate calculations. Furthermore, this method is independent of the battery management system, offering greater applicability and not being limited to batteries with such systems. It can obtain battery temperature distribution characteristics through data analysis and evaluate battery status, making it applicable to a wider range of scenarios.

[0049] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0050] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0051] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0052] Figure 1 This is a flowchart of the battery module temperature distribution inversion model construction method in Embodiment 1 of the present invention;

[0053] Figure 2 This is a flowchart of the battery module temperature distribution inversion model training in Embodiment 1 of the present invention;

[0054] Figure 3 This is a schematic diagram of the battery module temperature distribution inversion model construction device in Embodiment 1 of the present invention;

[0055] Figure 4 This is a flowchart of the battery module temperature distribution inversion method in Embodiment 2 of the present invention;

[0056] Figure 5 This is an example diagram of the temperature change curve in Embodiment 2 of the present invention;

[0057] Figure 6 This is a schematic diagram of the battery module temperature distribution inversion device in Embodiment 2 of the present invention. Detailed Implementation

[0058] As the energy storage device and power source for electric vehicles, the performance and lifespan of the power battery pack are greatly affected by temperature. The safe and stable operation of the battery pack is of great significance to the safety and reliability of the electric vehicle. Therefore, measuring the temperature of the electric vehicle's power battery pack and calculating its internal temperature distribution is crucial for assessing the electric vehicle's operating status and preventing thermal runaway caused by faults such as battery overheating and mechanical puncture. This is of great importance for the operation and maintenance of electric vehicle power batteries.

[0059] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0060] Existing battery overheating fault prediction methods rely on pure data-driven approaches, which make it difficult to obtain massive amounts of applicable and effective data in actual operation. They also ignore the battery operating principle and the physical relationships between its various parameters, resulting in poor prediction accuracy and data utilization efficiency. Furthermore, existing methods are not applicable in scenarios where only battery modules are available but a corresponding battery management system is lacking.

[0061] To address the problems of inaccurate predictions and poor data utilization efficiency in existing technologies, as well as their inapplicability in scenarios where only battery modules are available but a corresponding battery management system is lacking, embodiments of the present invention provide a method for inverting battery module temperature distribution and constructing an inversion model.

[0062] Example 1

[0063] Embodiment 1 of the present invention provides a method for constructing a battery module temperature distribution inversion model, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0064] Step S101: Obtain the temperature simulation results of the experimental battery module under multiple loading conditions using the finite element temperature simulation model of the experimental battery module.

[0065] Step S102: Based on the temperature simulation results, obtain a temperature distribution dataset including each set of loading conditions and corresponding temperature distribution data.

[0066] Step S103: Use the temperature distribution dataset to train the pre-established inversion model of power battery temperature to obtain the temperature distribution inversion model of the experimental battery module.

[0067] Preferably, in step S101 above, the finite element temperature simulation model includes an experimental battery pack geometric model and an electrothermal coupling physical field constructed based on the geometric model. The electrothermal coupling physical field includes at least an electric field distribution relationship, a temperature distribution relationship, and a total accumulated heat determined based on the relationship.

[0068] Preferably, a finite element temperature simulation model of the experimental battery module can be pre-established so that it can be used for temperature simulation and to collect temperature distribution data of the battery module under simulation conditions. The process of establishing the finite element temperature simulation model of the experimental battery module is as follows:

[0069] Based on the geometric features of the experimental battery module, a geometric model of the experimental battery module is constructed.

[0070] The grid data of the geometric model of the battery module is obtained by meshing the geometric model using a mesh partitioning method.

[0071] The Maxwell equations and solid heat transfer equations of the discretized experimental battery module are used to construct an electrothermal coupled physical field based on grid data, the discretized equations, and the discretized solid heat transfer equations.

[0072] In this embodiment, a finite element temperature simulation model of a single lithium-ion battery module is established using the finite element temperature simulation model establishment method described above for the experimental battery module:

[0073] Based on the geometric features of the lithium-ion battery module, a geometric model of the experimental battery module was constructed. The material and set parameters of the lithium-ion battery module setting area are shown in Table 1.

[0074] Table 1

[0075]

[0076] The geometric model of the lithium-ion battery module is meshed using a mesh analysis method to obtain the mesh data of the lithium-ion battery module geometric model. The mesh analysis method can be the ANSYS Maxwell adaptive mesh generation method, or other mesh analysis methods can be used according to actual needs. No specific restrictions are imposed here.

[0077] Maxwell's equations and solid-state heat transfer equations of a discretized lithium-ion battery module are used to construct an electrothermal coupled physical field based on the mesh data of the geometric model of the lithium-ion battery module, the discretized equations, and the discretized solid-state heat transfer equations. The electrothermal coupled physical field constructed based on the geometric model can reflect the physical laws of the experimental battery pack operation and the battery's operating state. Finally, a finite element temperature simulation model of the experimental battery module, including the geometric model and the electrothermal coupled physical field, is established. This finite element temperature simulation model can take into account the physical relationships between various battery parameters, so that the temperature distribution data obtained based on the simulation model is closer to the real data of the battery during operation.

[0078] Preferably, the constructed electrothermal coupling physical field includes at least the electric field distribution relationship, the temperature distribution relationship, and the total accumulated heat determined based on the relationship.

[0079] Optionally, the relationship between ion concentration migration and current density can be added to the above electrothermal coupling physical field to reflect the strength of the electric field. This allows the constructed electrothermal coupling physical field to more accurately and realistically reflect the operating state of the lithium-ion battery module and generate more accurate temperature data in the simulation.

[0080] The coupling relationship between the solid active material phase and the electrolyte phase in a porous electrode can be established using the following formula:

[0081]

[0082]

[0083] In the formula j loc The local current density is represented by j0; the exchange current density is c. surf This represents the insertion exchange current on the surface of the active material; α a α c Indicates the transfer coefficient at the anode and cathode; This represents the solid-state potential in the battery; E represents the liquid phase potential in the battery. OCP Represents the open-circuit voltage of the battery; F represents the Faraday constant; R represents the gas constant; a represents the specific area of ​​the active material; j L J is the current density in the liquid phase. n The current density represents the normal phase current density, and T is the temperature.

[0084] The migration relationship of lithium ion concentration over time can be established using the following formula:

[0085]

[0086] Where c L Represents lithium ion concentration; t represents time; V m LiPF6 ε represents the molar volume of LiPF6; L β represents the dielectric constant of the liquid phase in the battery. L D represents the Bragman constant in the liquid phase; L Indicates the diffusion coefficient in the liquid phase; The value is (0.28687×(c)). L / 1000)2+0.74678×(c L / 1000)+0.44103) / (1-t Li+ solv );c solv This indicates the lithium-ion concentration relative to the solvent; c L tot This represents the total lithium-ion concentration; t Li+ solv j represents the lithium-ion transference number relative to the solvent. L is the current density in the liquid phase; T is the temperature.

[0087] The mass transport of lithium in the active material particles of the positive and negative electrodes is assumed to be diffusion along the particle radius / thickness and is described by Fick's second law.

[0088] Preferably, the heat in the local area of ​​the battery module is determined based on the above-mentioned electric field distribution relationship and temperature distribution relationship, and the internal heat generation power of the battery is calculated according to the following formula: q cc / sc =(js) 2 / σ, where j S σ is the current density; σ is the electron conductivity, and q is the current density. cc / scThis refers to the heating power of local Joule heating; the total accumulated heat inside the battery is determined by calculating the relationship between the heating power and the time change during battery operation. Further calculation of the total accumulated heat inside the battery within a physical field incorporating electric and temperature distribution relationships allows the final finite element temperature simulation model to more realistically reflect the battery's operating state, resulting in more accurate data.

[0089] Preferably, in step S101 above, the finite element temperature simulation model of the experimental battery module is used to simulate the temperature under multiple loading conditions of the experimental battery module under preset operating conditions, and the temperature simulation results are obtained, including:

[0090] Set the operating conditions of the experimental battery module and multiple sets of loading conditions under different operating conditions;

[0091] The finite element temperature simulation model of the experimental battery module was used to simulate the battery's operating state under different loading conditions; and the surface temperature and internal temperature of the experimental battery module under loading conditions were collected to obtain temperature distribution data of the experimental battery module under multiple sets of loading conditions under preset working conditions.

[0092] The preset operating conditions include at least one of normal operating conditions and fault operating conditions; the loading conditions include at least one of the following: ambient temperature of the simulation model, internal short-circuit fault conditions, and per-unit value of charging power.

[0093] The finite element temperature simulation model of the battery module is used to simulate the battery's operating state under different loading conditions, and temperature distribution data based on the simulation model is obtained. The finite element temperature simulation model can obtain a large amount of accurate temperature data, effectively improving the difficulty of obtaining large amounts of applicable and effective data in traditional methods. The loading conditions include at least one of the following: ambient temperature of the simulation model, internal short-circuit fault condition, and per-unit charging power. The internal short-circuit fault condition can be achieved by setting the internal short-circuit fault radius. When the internal short-circuit fault radius is 0, it indicates that the battery module is operating under normal conditions; when the internal short-circuit fault radius is any other value, it is considered a fault condition. The per-unit charging power value is determined by the ratio between the test charging power of the battery module and the full-load charging power under standard operating conditions.

[0094] Preferably, in step S102 above, a temperature distribution dataset is obtained based on the temperature simulation results, including each set of loading conditions and corresponding temperature distribution data, comprising:

[0095] Acquire temperature distribution data for the experimental battery modules under each loading condition; the temperature distribution data includes the internal temperature of the experimental battery modules and their corresponding surface temperatures;

[0096] Based on each set of loading conditions and the corresponding temperature distribution data, multiple sets of first temperature distribution arrays are established.

[0097] The experimental battery module temperature distribution dataset is constructed based on multiple sets of first temperature distribution arrays.

[0098] Each set of loading conditions includes, but is not limited to, the ambient temperature of the simulation model, the internal short-circuit fault condition, and the per-unit value of the charging power. By changing the values ​​of the loading conditions, multiple sets of loading conditions are obtained. The established first temperature distribution array includes a temperature distribution dataset containing each set of loading conditions and the corresponding temperature distribution data. For example, a set of first temperature distribution arrays can be represented in the following form [T0, ..., T n [T = 0℃, R = 0um, C_rate = 1], this array represents the temperature distribution data of the battery module obtained under the loading conditions of ambient temperature of 0℃, internal short-circuit fault radius of 0um, and charging power per unit value of 1. The operating conditions under these conditions are normal operating conditions, where T0~T n This represents the internal temperature of the battery module at the selected location and the corresponding surface temperature of the battery; the first temperature distribution array [T0, ..., T...] n [T = 0℃, R = 10um, C_rate = 1], this array represents the battery module temperature distribution data obtained under the loading conditions of ambient temperature of 0℃, internal short-circuit fault radius of 10um, and charging power per unit value of 1. This operating condition is the fault condition. The ambient temperature, internal short-circuit fault condition, and charging power per unit value of the simulation model can be changed as needed, and T0~T10 ... n In this embodiment, the ambient temperature can be 0℃, 20℃, 40℃, ..., the internal short-circuit fault radius can be 0um, 10um, 20um, ..., and the per-unit value of charging power can be 1, 3, 5, ... Of course, the values ​​of the loading conditions can also be specified according to actual needs. By changing the values ​​of the loading conditions of the experimental battery module, multiple sets of temperature distribution data under loading conditions can be obtained.

[0099] Preferably, step S102 above further includes:

[0100] The temperatures in each group of the first temperature distribution array in the temperature distribution dataset are normalized using the following formula: in For the normalized temperature data, T i Temperature data for the selected temperature points; T max With T min These are the maximum and minimum temperatures in the first temperature distribution array, respectively. The purpose of normalization is to unify the numerical ranges of different features to the same scale, thereby improving the training effect and prediction accuracy of the model.

[0101] Preferably, in step S103 above, the temperature distribution dataset is used to train the pre-established inversion model of the power battery temperature to obtain the temperature distribution inversion model of the experimental battery module, including:

[0102] According to the accuracy requirements, the temperature distribution dataset is divided into N1 groups of first temperature distribution subarrays under normal operating conditions and N2 groups of first temperature distribution subarrays under fault operating conditions. Each first temperature distribution subarray is represented by a feature matrix consisting of M data points.

[0103] The first temperature distribution subarray of group N1 under normal operating conditions and the first temperature distribution subarray of group N2 under fault operating conditions are respectively input into the inversion model of power battery temperature pre-built based on backpropagation neural network. The internal temperature prediction value of the experimental battery module is output. Based on the internal temperature prediction value and the internal temperature of the experimental battery module included in the temperature distribution dataset, the parameters of the inversion model are tuned, and the model training process is repeated until the model meets the preset convergence condition. The trained power battery temperature inversion model is obtained as the temperature distribution inversion model of the experimental battery module.

[0104] For example, the temperature distribution subarrays can be divided into N1 group for normal operating conditions and N2 group for fault operating conditions according to time precision. Each group of first temperature distribution subarrays is represented by a feature matrix composed of M data points. In this embodiment, a feature matrix composed of 11 data points is used, including temperature data at 8 temperature points and data for 3 loading conditions, for example, in the following form: [T0, ..., T7, T = 0℃, R = 0um, C_rate = 1], [T0, ..., T7, T = 20℃, R = 10um, C_rate = 3], [T0, ..., T7, T = 40℃, R = 20um, C_rate = 5], ... According to the value of the radius R of the internal short circuit fault, the N1 subarray under normal operating conditions and the subarray under fault conditions in the first temperature distribution array can be determined. Since batteries typically operate in only two states—normal operation and fault—it is necessary to divide the temperature distribution data under different operating conditions. The temperature distribution data under normal and fault conditions should be input into the power battery temperature propagation model for training. This will result in a more accurate model, making the output data closer to the true value, thereby improving the accuracy of temperature prediction.

[0105] Preferably, based on the predicted internal temperature and the internal temperature of the experimental battery module included in the temperature distribution subarray, the inversion model is parameter-tuned, and the model training process is repeated until the model meets the preset convergence conditions, including:

[0106] The error between the predicted internal temperature output by the inversion model and the measured internal temperature of the experimental battery module included in the temperature distribution dataset is calculated based on the pre-constructed error function. The partial derivative of the error with respect to the initial weights of the inversion model is also calculated to obtain the weight adjustment gradient of the inversion model. The weights are then updated based on the weight adjustment gradient. The error function of the inversion model is as follows: Among them O i t represents the predicted internal temperature output by the inversion model. i The measured internal temperature values ​​of the experimental battery module obtained from the simulation model;

[0107] The process of model training and error calculation is repeated multiple times until the error reaches the preset requirement, thus completing the training.

[0108] During model training, given the input data and initial weights, the model can output a predicted internal temperature. However, whether this output value is truly effective needs to be determined by calculating the error function between the predicted internal temperature and the measured internal temperature of the experimental battery module obtained from the simulation model. If the error is large, the weights need to be adjusted, and the training process repeated. This continuous adjustment of the calculation error and weights ultimately minimizes the error and improves the model's fit. The method for adjusting the weights is to calculate the partial derivative of the error with respect to the weights, obtain the adjustment gradient, and continuously update the weights based on the adjustment gradient during the iteration process. In this embodiment, the temperature distribution inversion model of the power battery is trained based on the temperature distribution data obtained from the lithium battery temperature simulation model. Therefore, the trained power battery temperature inversion model is used as the temperature distribution inversion model of the lithium battery module, and then the temperature distribution inversion model of the lithium battery module is applied to practical applications.

[0109] Figure 2 The flowchart for training the temperature distribution inversion model of the battery module is as follows: First, set the values ​​of different loading conditions, then input the loading conditions into the finite element temperature simulation model to perform temperature simulation, obtain the temperature distribution data, and then use the loading conditions and the corresponding temperature distribution data to form a sample set for model training. Finally, through multiple iterations of training, until the model meets the training requirements, the trained temperature distribution inversion model of the experimental battery module is obtained.

[0110] Based on the same inventive concept, embodiments of the present invention also provide a battery module temperature distribution inversion model construction device. This device can be installed in a device capable of processing computer instructions, and its structure is as follows: Figure 3 As shown, it includes:

[0111] Data acquisition module 11 is used to acquire temperature simulation results of multiple loading conditions of the experimental battery module under preset working conditions using a finite element temperature simulation model of the experimental battery module. The finite element temperature simulation model includes a geometric model of the experimental battery pack and an electrothermal coupling physical field constructed based on the geometric model. The electrothermal coupling physical field includes at least an electric field distribution relationship, a temperature distribution relationship, and a total accumulated heat determined based on the relationship. Based on the temperature simulation results, a temperature distribution dataset including each loading condition and the corresponding temperature distribution data is obtained. The temperature distribution data includes the surface temperature and internal temperature of the battery module.

[0112] Model building module 12 is used to train a pre-established inversion model of power battery temperature using a temperature distribution dataset to obtain a temperature distribution inversion model of the experimental battery module.

[0113] Preferably, the model building module 12 is specifically used to input the first temperature distribution subarray of the N1 group under normal operating conditions and the first temperature distribution subarray of the N2 group under fault operating conditions from the temperature distribution dataset into the inversion model of power battery temperature pre-built based on the backpropagation neural network, output the predicted internal temperature value of the experimental battery module, adjust the parameters of the inversion model based on the predicted internal temperature value and the internal temperature of the experimental battery module included in the temperature distribution dataset, and repeat the model training process until the model meets the preset convergence condition, so as to obtain the trained inversion model of power battery temperature as the temperature distribution inversion model of the experimental battery module.

[0114] Regarding the battery module temperature distribution inversion model construction device in the above embodiments, the specific way in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0115] The method and apparatus described in this embodiment establish a geometric model based on the geometric characteristics of the experimental battery pack, and construct an electrothermal coupling physical field based on the geometric model to reflect the physical laws of operation and the battery's operating state. Finally, a finite element temperature simulation model of the experimental battery module, including the geometric model and the electrothermal coupling physical field, is established. The operating state of the battery pack is simulated based on the constructed simulation model, and surface and internal temperature data are collected during the simulation process. The obtained temperature data reflects the physical relationships between various battery parameters, more closely resembling the actual situation of the experimental battery pack. Based on the obtained surface and internal temperature data, the inversion model of the power battery temperature is trained. This trained model can better predict the internal temperature of the battery based on the battery surface temperature, obtaining more accurate temperature prediction results and more timely and accurate prediction of potential overheating states of the battery.

[0116] Example 2

[0117] Embodiment 2 of the present invention provides a specific implementation process of the battery module temperature distribution inversion method, the process of which is as follows: Figure 4 As shown, it includes the following steps:

[0118] Step S201: Collect surface temperature distribution data of the experimental battery module to be predicted under the preset working conditions;

[0119] Step S202: Input the loading conditions of the preset working condition and the corresponding surface temperature distribution data into the temperature distribution inversion model of the trained experimental battery module, and output the internal temperature of the experimental battery module under the preset working condition.

[0120] Preferably, in step S201 above, collecting surface temperature distribution data under preset loading conditions of the experimental battery module includes:

[0121] Under preset operating conditions and preset loading conditions, the temperature at points on the surface of the battery module is obtained by temperature sensors arranged in the finite element temperature simulation model of the experimental battery module, thus obtaining the surface temperature distribution data of the experimental battery module under preset operating conditions and preset loading conditions.

[0122] Preferably, after obtaining the surface temperature distribution data under the preset loading conditions of the experimental battery module, a second temperature distribution array is established based on the loading conditions of the preset working conditions and the corresponding surface temperature distribution data.

[0123] Preferably, the temperature sensors arranged in the finite element temperature simulation model of the experimental battery module include: A temperature sensors arranged at the positive electrode of the battery, B temperature sensors arranged at the negative electrode of the battery, and C temperature sensors arranged on the side of the battery, with the C temperature sensors arranged on the side of the battery being evenly distributed.

[0124] For example, to collect surface temperature distribution data of a lithium battery under normal operating conditions, the loading conditions can be set to T = 20℃, R = 0um, and C_rate = 1. The temperature of a region on the battery surface can be measured using one temperature sensor placed at the positive electrode, one temperature sensor placed at the negative electrode, and six temperature sensors placed on the side of the battery. The six temperature sensors on the side of the battery are evenly distributed, and the spacing between the temperature sensors can be set, for example, to 776um. The measured temperatures are labeled as T1-T8, constructing a second temperature distribution array [T1, T2, ..., T8, T = 20℃, R = 0um, C_rate, R = 1].

[0125] Preferably, in step S202 above, the loading conditions of the preset working condition and the corresponding surface temperature distribution data are input into the trained temperature distribution inversion model of the experimental battery module, and the internal temperature of the experimental battery module under the preset working condition is output. The temperature distribution inversion model of the experimental battery module is obtained by the battery module temperature distribution inversion model construction method in Example 1. Using the trained experimental battery module temperature distribution inversion model, the internal temperature of the battery can be predicted based on the battery surface temperature. It can obtain the battery temperature distribution characteristics and evaluate the battery state through data analysis without relying on the battery management system, and has a wider range of applicable scenarios. This embodiment obtains the temperature distribution inversion model of a lithium battery module; the temperature distribution inversion models of other battery modules can be retrained and obtained according to actual conditions.

[0126] Preferably, step S202 further includes:

[0127] The internal temperature of the experimental battery module under the preset operating conditions is fitted to obtain the temperature change curve.

[0128] Over time, the temperature of a lithium battery changes continuously, so the temperature measured at each point on the battery surface is constantly changing. The battery generates heat internally, which is why surface temperatures can be measured; therefore, the internal temperature is also constantly changing. By fitting the internal temperatures of the experimental battery module under preset operating conditions at multiple time points, a temperature change curve can be obtained. This curve can determine when the temperature will reach its maximum. When the internal temperature of the battery is about to reach its maximum, a high-temperature warning is issued, thereby preventing thermal runaway and improving the safety and reliability of electric vehicles.

[0129] Figure 5 This is an example of a temperature change curve. The curve is generated under normal operating conditions with an ambient temperature of 20°C and a charging power per unit value of 1. Multiple sets of second temperature distribution arrays are obtained and input into the lithium battery module temperature distribution inversion model. The model outputs the internal temperature of the lithium battery at multiple time points. Based on the fitted temperature change curve at these multiple time points, the temperature change is calculated... Figure 5 It can be seen that under normal operating conditions, when the ambient temperature is 20℃ and the per-unit charging power is 1, the temperature of the lithium battery module gradually increases with time, and then stabilizes after reaching a maximum internal temperature of 58℃ in 0.008s.

[0130] Steps S201 to S202 realize the prediction of the internal temperature of the experimental battery pack based on the trained experimental battery module temperature distribution inversion model and the surface temperature distribution data of the experimental battery pack under the preset working conditions, thereby obtaining accurate temperature prediction results.

[0131] Based on the same inventive concept, embodiments of the present invention also provide a battery module temperature distribution inversion device. This device can be installed in a device with computer instruction processing capabilities, and its structure is as follows: Figure 6 As shown, it includes:

[0132] The data acquisition module 21 is used to collect surface temperature distribution data of the experimental battery module to be predicted under the preset working conditions.

[0133] The temperature prediction module 22 is used to input the loading conditions of the preset working conditions and the corresponding surface temperature distribution data into the trained temperature distribution inversion model of the experimental battery module, and output the internal temperature of the experimental battery module under the preset working conditions; the temperature distribution inversion model is trained using the battery module temperature distribution inversion model construction method.

[0134] Preferably, the temperature prediction module 22 is also used to fit the internal temperature of the experimental battery module under the preset operating conditions to obtain the temperature change curve.

[0135] Regarding the pool module temperature distribution inversion device in the above embodiments, the specific way in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0136] The method and apparatus described in this invention predict the internal temperature of the experimental battery pack based on the trained battery module temperature distribution inversion model and the surface temperature distribution data of the experimental battery pack under preset loading conditions. This allows for accurate temperature prediction results, timely prediction of internal temperature changes in the battery, and timely warnings and appropriate actions when overheating is possible. This helps avoid accidents caused by overheating, improves the safety of the power battery, and ultimately ensures the safe use of electric vehicles.

[0137] Unless otherwise specifically stated, terms such as processing, calculation, operation, determination, display, etc., may refer to the actions and / or processes of one or more processing or computing systems or similar devices that represent the manipulation and conversion of data representing physical (e.g., electronic) quantities within the registers or memory of the processing system into other data similarly representing physical quantities within the memory, registers, or other such information storage, transmission, or display devices of the processing system. Information and signals can be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the above description can be represented by voltage, current, electromagnetic waves, magnetic fields or particles, light fields or particles, or any combination thereof.

[0138] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to the specific order or hierarchy described.

[0139] In the detailed description above, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features in a single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, with each claim representing a separate preferred embodiment of the invention.

[0140] Those skilled in the art will also understand that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments herein can be implemented as electronic hardware, computer software, or a combination thereof. To clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the described functionality in alternative ways for each specific application; however, such implementation decisions should not be construed as departing from the scope of this disclosure.

[0141] The steps of the methods or algorithms described in conjunction with the embodiments herein can be directly embodied in hardware, software modules executed by a processor, or a combination thereof. The software modules can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in a user terminal. Alternatively, the processor and storage medium can exist as discrete components in the user terminal.

[0142] For software implementation, the techniques described in this application can be implemented using modules (e.g., procedures, functions, etc.) that perform the functions described in this application. This software code can be stored in memory units and executed by a processor. The memory units can be implemented within the processor or outside the processor; in the latter case, they are communicatively coupled to the processor via various means, as is well known in the art.

[0143] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

Claims

1. A method for constructing a battery module temperature distribution inversion model, characterized in that, include: The temperature simulation results were obtained by using the finite element temperature simulation model of the experimental battery module to simulate the temperature under multiple loading conditions of the experimental battery module under preset working conditions. The finite element temperature simulation model includes a geometric model of the experimental battery pack and an electrothermal coupled physical field constructed based on the geometric model. The electrothermal coupled physical field includes at least an electric field distribution relationship, a temperature distribution relationship, and a total accumulated heat determined based on the relationship. Based on the temperature simulation results, a temperature distribution dataset is obtained, which includes each set of loading conditions and the corresponding temperature distribution data; the temperature distribution data includes the surface temperature and internal temperature of the battery module. The temperature distribution dataset is used to train a pre-established inversion model of power battery temperature to obtain a temperature distribution inversion model of the experimental battery module. The inversion model can predict the internal temperature of the battery based on the surface temperature of the battery.

2. The method as described in claim 1, characterized in that, The process of establishing the finite element temperature simulation model of the experimental battery module is as follows: Based on the geometric features of the experimental battery module, a geometric model of the experimental battery module is constructed. The grid data of the geometric model of the battery module is obtained by meshing the geometric model using a mesh partitioning method. The Maxwell equations and solid heat transfer equations of the discretized experimental battery module are used to construct an electrothermal coupled physical field based on the grid data, the discretized equations, and the discretized solid heat transfer equations.

3. The method as described in claim 1, characterized in that, The finite element temperature simulation model of the experimental battery module is used to simulate the temperature under multiple loading conditions of the experimental battery module under preset operating conditions, and the temperature simulation results are obtained, including: Set the operating conditions of the experimental battery module and multiple sets of loading conditions under different operating conditions; The finite element temperature simulation model of the experimental battery module was used to simulate the battery's operating state under different loading conditions; and the surface temperature and internal temperature of the experimental battery module under loading conditions were collected to obtain temperature distribution data of the experimental battery module under multiple sets of loading conditions under preset working conditions. The preset operating conditions include at least one of normal operating conditions and fault operating conditions; the loading conditions include at least one of the following: ambient temperature of the simulation model, internal short-circuit fault conditions, and per-unit value of charging power.

4. The method as described in claim 1, characterized in that, Based on the temperature simulation results, a temperature distribution dataset was obtained, including each set of loading conditions and the corresponding temperature distribution data; including: Acquire temperature distribution data for the experimental battery modules under each loading condition; the temperature distribution data includes the internal temperature of the experimental battery modules and their corresponding surface temperatures; Based on each set of loading conditions and the corresponding temperature distribution data, multiple sets of first temperature distribution arrays are established. The experimental battery module temperature distribution dataset is constructed based on multiple sets of first temperature distribution arrays.

5. The method as described in claim 4, characterized in that, Also includes: The temperatures in each group of the first temperature distribution array in the temperature distribution dataset are normalized using the following formula: ,in The temperature data is normalized. T i Temperature data for the selected temperature points; T max and T min These are the maximum and minimum temperatures in the first temperature distribution array, respectively.

6. The method as described in claim 1, characterized in that, The temperature distribution inversion model of the experimental battery module was trained using a temperature distribution dataset to obtain the temperature distribution inversion model of the experimental battery module, including: According to the accuracy requirements, the temperature distribution dataset is divided into N1 groups of first temperature distribution subarrays under normal operating conditions and N2 groups of first temperature distribution subarrays under fault operating conditions. Each first temperature distribution subarray is represented by a feature matrix consisting of M data points. The first temperature distribution subarray of group N1 under normal operating conditions and the first temperature distribution subarray of group N2 under fault operating conditions are respectively input into the inversion model of power battery temperature pre-built based on backpropagation neural network. The internal temperature prediction value of the experimental battery module is output. Based on the internal temperature prediction value and the internal temperature of the experimental battery module included in the temperature distribution dataset, the parameters of the inversion model are tuned, and the model training process is repeated until the model meets the preset convergence condition. The trained power battery temperature inversion model is obtained as the temperature distribution inversion model of the experimental battery module.

7. The method as described in claim 6, characterized in that, Based on the predicted internal temperature and the internal temperatures of the experimental battery modules included in the temperature distribution subarray, the inversion model is parameter-tuned, and the model training process is repeated until the model meets the preset convergence conditions, including: The error between the predicted internal temperature output by the inversion model and the measured internal temperature of the experimental battery module included in the temperature distribution dataset is calculated based on a pre-constructed error function. The partial derivative of the error with respect to the initial weights of the inversion model is also calculated to obtain the weight adjustment gradient of the inversion model. The weights are then updated based on this gradient. The error function of the inversion model is as follows: ,in O i The predicted internal temperature output by the inversion model. t i The measured internal temperature values ​​of the experimental battery module obtained from the simulation model; The process of model training and error calculation is repeated multiple times until the error reaches the preset requirement, thus completing the training.

8. A method for inverting the temperature distribution of a battery module, characterized in that, include: Collect surface temperature distribution data of the experimental battery module to be predicted under the preset working conditions loading conditions, input the preset working conditions loading conditions and the corresponding surface temperature distribution data into the trained temperature distribution inversion model of the experimental battery module, and output the internal temperature of the experimental battery module under the preset working conditions. The temperature distribution inversion model is trained using the battery module temperature distribution inversion model construction method described in any one of claims 1-7.

9. The method as described in claim 8, characterized in that, Surface temperature distribution data of the experimental battery module under preset loading conditions were collected, including: Under preset operating conditions and preset loading conditions, the temperature at points on the surface of the battery module is obtained by temperature sensors arranged in the finite element temperature simulation model of the experimental battery module, thus obtaining the surface temperature distribution data of the experimental battery module under preset operating conditions and preset loading conditions.

10. The method as described in claim 9, characterized in that, The temperature sensors arranged in the finite element temperature simulation model of the experimental battery module include: A temperature sensor A is arranged at the positive terminal of the battery, B temperature sensor B is arranged at the negative terminal of the battery, and C temperature sensor C is arranged on the side of the battery. The C temperature sensors arranged on the side of the battery are evenly distributed.

11. A device for constructing a battery module temperature distribution inversion model, characterized in that, include: The data acquisition module is used to acquire the temperature simulation results of the experimental battery module under multiple loading conditions based on the finite element temperature simulation model of the experimental battery module. The finite element temperature simulation model includes the geometric model of the experimental battery module and the electrothermal coupling physical field constructed based on the geometric model. The electrothermal coupling physical field includes at least the electric field distribution relationship, the temperature distribution relationship, and the total accumulated heat determined based on the relationship. Based on the temperature simulation results, a temperature distribution dataset is obtained, which includes each set of loading conditions and the corresponding temperature distribution data; the temperature distribution data includes the surface temperature and internal temperature of the battery module. The model building module is used to train a pre-established inversion model of power battery temperature using the temperature distribution dataset to obtain a temperature distribution inversion model of the experimental battery module. The inversion model can predict the internal temperature of the battery based on the battery surface temperature.

12. A battery module temperature distribution inversion device, characterized in that, include: The data acquisition module is used to collect surface temperature distribution data of the experimental battery module to be predicted under preset loading conditions. The temperature prediction module is used to input the loading conditions of the preset working conditions and the corresponding surface temperature distribution data into the trained temperature distribution inversion model of the experimental battery module, and output the internal temperature of the experimental battery module under the preset working conditions; the temperature distribution inversion model is trained using the battery module temperature distribution inversion model construction method described in any one of claims 1-7.

13. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions, which, when executed by a processor, implement the battery module temperature distribution inversion model construction method according to any one of claims 1-7 or the battery module temperature distribution inversion method according to any one of claims 8-10.

14. A computer device, characterized in that, include: The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the battery module temperature distribution inversion model construction method according to any one of claims 1-7 or the battery module temperature distribution inversion method according to any one of claims 8-10.

Images