Lithium ion battery thermal runaway prediction analysis method, device, equipment and medium

Abstract

The invention discloses a lithium ion battery thermal runaway prediction analysis method, apparatus and device, and a medium. The method comprises the steps of obtaining surface temperature data of a lithium ion battery; a battery thermal simulation model is adopted, and the internal temperature of the battery is determined and estimated according to the surface temperature data; the battery thermal simulation model is a physical simulation model which is determined by triggering a battery thermal runaway test and based on signal data acquired by temperature sensors arranged outside a lithium ion battery packaging mode and at the central position of the surface of a roll core in the battery and is used for representing a mapping relationship between internal and external temperature signals of the battery; determining the internal temperature of the predicted battery according to the surface temperature data by adopting a temperature prediction model; and performing fusion processing according to the estimated internal temperature of the battery and the predicted internal temperature of the battery to obtain an internal temperature estimation result of the battery for representing the thermal state of the lithium ion battery. The invention aims to realize the prediction analysis of the thermal runaway of the lithium ion battery and improve the prediction precision.

Description

Technical Field

[0001] The present application relates to the field of battery thermal runaway prediction and analysis, and in particular to a lithium-ion battery thermal runaway prediction and analysis method, device, equipment and medium. Background Art

[0002] With the widespread application of lithium-ion batteries in electric vehicles, consumer electronics and energy storage systems, how to effectively monitor and manage the thermal state of the battery and prevent thermal runaway has become a key issue in battery safety and reliability design.

[0003] Lithium-ion batteries may experience thermal runaway under overcharge, over-discharge, battery damage or extreme working conditions. Thermal runaway refers to a self-sustaining thermal reaction inside the battery, which causes a sharp rise in temperature and may eventually cause combustion or explosion. This situation will not only cause battery damage, but may also bring serious safety hazards, especially in electric vehicles or large energy storage devices, where the consequences of thermal runaway are very serious.

[0004] The main factors leading to thermal runaway include: Internal short circuit: Due to damage to the diaphragm, decomposition of the electrolyte, etc., the battery internal short circuit occurs, generating a lot of heat.

[0005] Overcharge or over-discharge: Improper charging or discharging operations may cause abnormal electrochemical reactions inside the battery and generate a lot of heat.

[0006] Extreme external temperature: Too high or too low external temperature may affect the internal temperature of the battery and trigger thermal runaway.

[0007] To prevent thermal runaway, the battery temperature must be accurately monitored, especially the temperature changes inside the battery, because rising internal temperature is often an early sign of thermal runaway.

[0008] Traditional battery temperature monitoring mainly relies on external sensors, such as temperature sensors installed near the battery housing or electrodes. However, external temperature sensors can only monitor temperature changes outside the battery, and it is difficult to directly obtain internal temperature, especially the temperature of the battery cell. The temperature of the battery cell is often the source of thermal runaway. Therefore, relying solely on external temperature monitoring may not detect the problem in time before thermal runaway occurs. In addition, it is also challenging to install sensors directly inside the battery. The arrangement of internal sensors will affect the battery structure, and the complex battery structure limits the number and accuracy of internal sensors.

[0009] Therefore, how to achieve prediction and analysis of thermal runaway of lithium-ion batteries is crucial. Summary of the invention

[0010] The purpose of the present application is to provide a method, device, equipment and medium for predicting and analyzing thermal runaway of a lithium-ion battery, which can realize predictive analysis of thermal runaway of a lithium-ion battery and improve prediction accuracy.

[0011] To achieve the above objectives, this application provides the following solutions: In a first aspect, the present application provides a method for predicting and analyzing thermal runaway of a lithium-ion battery, comprising: Obtain surface temperature data of lithium-ion batteries; A battery thermal simulation model is used to determine the estimated internal temperature of the battery according to the surface temperature data; the battery thermal simulation model is a physical simulation model for characterizing the mapping relationship between the internal and external temperature signals of the battery, determined by triggering a battery thermal runaway test and based on signal data collected by temperature sensors arranged outside the lithium-ion battery packaging and at the center of the surface of the internal winding core of the battery; Using a temperature prediction model to predict the internal temperature of the battery according to the surface temperature data; the temperature prediction model is determined by a machine learning method; A fusion process is performed based on the estimated battery internal temperature and the predicted battery internal temperature to obtain a battery internal temperature estimation result; the battery internal temperature estimation result is used to characterize the thermal state of the lithium-ion battery.

[0012] Optionally, the battery thermal simulation model includes: a heat generation model and a heat transfer model; The heat generation model is determined based on the ohmic heat generated by the battery resistance effect, the polarization heat generated by the deviation of the electrode potential from the equilibrium potential during the battery charging and discharging process, and the reaction heat generated by the electrochemical reaction between the electrode material and the electrolyte during the charging and discharging process; The heat transfer model is determined based on the heat generated by heat conduction, convection, radiation and electrochemical reaction of the lithium-ion battery.

[0013] Optionally, the calculation formula of the heat generation model is: ; ; ; ; in, is the mathematical expression of the heat generation model; for ohmic heat; is the heat of reaction; is polarization heat; is the working current; is the internal resistance of the battery; is the electrochemical reaction equivalent number of the battery; is the Faraday constant; is the open circuit voltage; is temperature; is the partial derivative of the open circuit voltage with respect to temperature; is the voltage drop caused by the resistor; is the voltage drop caused by the difference in ion concentration; is the voltage drop caused by the electrode activation energy.

[0014] Optionally, the method for determining the temperature prediction model specifically includes: Acquire training data; the training data includes surface measurement temperature data of the lithium-ion battery and corresponding label data; the label data includes: internal measurement temperature data; Construct an initial decision tree; Input the training data into the initial decision tree, and perform iterative training guidance on the initial decision tree based on the gradient and the second-order gradient to obtain a decision tree; wherein the gradient is the first-order derivative of the loss function, and the second-order gradient is the second-order derivative of the loss function; and the loss function is determined based on the label data and the data output by the decision tree; Performing weighted sum processing on the decision tree based on the regularization term to obtain the temperature prediction model; Among them, for any iteration: Select the split point based on the gain function and determine the split node; A tree structure is generated according to the split node, and a pruning judgment is performed based on an evaluation condition to obtain a first judgment result; the evaluation condition is that the gain of the split node is less than a preset threshold; If the first judgment result is yes, the tree structure is pruned, and the step of "selecting a split point based on a gain function and determining a split node" is returned; If the first judgment result is no, the tree structure is scaled according to the learning rate, and it is determined whether a stopping condition is met to obtain a second judgment result; the stopping condition is that the loss function is within a preset range or the number of iterations reaches a set maximum round; If the second judgment result is yes, the tree structure corresponding to the current number of iterations is used as the decision tree; If the second judgment result is no, the process returns to “selecting split points based on the gain function and determining split nodes”.

[0015] Optionally, the calculation formula of the battery internal temperature estimation result is: ; in, is the estimated result of the internal temperature of the battery; is the first weight; To estimate the internal temperature of the battery; is the second weight; To predict the internal temperature of the battery.

[0016] Optionally, the battery thermal runaway test includes: puncture, overheating and overcharging.

[0017] In a second aspect, the present application provides a lithium-ion battery thermal runaway prediction and analysis device, comprising: A data acquisition module, used to acquire surface temperature data of the lithium-ion battery; a temperature estimation module, for determining an estimated internal temperature of the battery according to the surface temperature data using a battery thermal simulation model; the battery thermal simulation model is a physical simulation model for characterizing a mapping relationship between internal and external temperature signals of the battery, determined by triggering a battery thermal runaway test and based on signal data collected by temperature sensors disposed outside the lithium-ion battery packaging and at the center of the surface of the battery internal winding core; A temperature prediction module, used to use a temperature prediction model to determine and predict the internal temperature of the battery according to the surface temperature data; the temperature prediction model is determined by a machine learning method; The fusion processing module is used to perform fusion processing according to the estimated battery internal temperature and the predicted battery internal temperature to obtain a battery internal temperature estimation result; the battery internal temperature estimation result is used to characterize the thermal state of the lithium-ion battery.

[0018] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method for predicting and analyzing thermal runaway of a lithium-ion battery.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for predicting and analyzing thermal runaway of a lithium-ion battery.

[0020] According to the specific embodiments provided in this application, this application has the following technical effects: The present application provides a method, device, equipment and medium for predicting and analyzing thermal runaway of lithium-ion batteries, by acquiring the surface temperature data of lithium-ion batteries; using a battery thermal simulation model to determine the estimated internal temperature of the battery based on the surface temperature data; the battery thermal simulation model is a physical simulation model that is determined by triggering a battery thermal runaway test, based on the signal data collected by temperature sensors set at the center of the surface of the battery internal winding core outside the lithium-ion battery packaging method and inside the battery, to characterize the mapping relationship between the temperature signals inside and outside the battery; using a temperature prediction model to determine the predicted internal temperature of the battery based on the surface temperature data; performing fusion processing based on the estimated internal temperature of the battery and the predicted internal temperature of the battery to obtain an estimated result of the internal temperature of the battery, which is used to characterize the thermal state of the lithium-ion battery. The present application realizes predictive analysis of thermal runaway of lithium-ion batteries by combining a battery thermal simulation model with a temperature prediction model, and fusion processing of the temperatures corresponding to the two models. Since the temperature prediction model is determined based on a machine learning method, the prediction accuracy can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 This is a flow chart of the method for predicting and analyzing thermal runaway of a lithium-ion battery mentioned in an embodiment of the present application; Figure 2 A flowchart of the technical concept of the lithium-ion battery thermal runaway prediction and analysis method provided in the embodiment of the present application; Figure 3 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] The development of simulation technology and machine learning has provided new solutions for battery temperature management. Thermal simulation technology is based on the physical modeling of heat transfer inside the battery and can predict the temperature changes in different parts of the battery. Machine learning trains the model through a large amount of experimental data and uses external sensor data to infer the internal temperature, thereby achieving high-precision temperature estimation.

[0025] Thermal simulation is a computational model based on the characteristics of battery materials and the principle of heat transfer. It takes into account the heat sources inside the battery (such as electrochemical reactions, ohmic heat, etc.) and the influence of the external thermal environment, and calculates the temperature changes of various parts inside the battery. Through the thermal simulation model, the thermal performance of the battery under different operating conditions can be predicted during the design phase. Machine learning technology trains the model through a large amount of experimental data, especially establishing a mapping relationship between temperature sensor data and the actual internal temperature of the battery. Based on the data input from external sensors, the machine learning model can infer the internal temperature distribution. This data-driven model can adapt to different battery types and usage environments, and is highly flexible and scalable.

[0026] With the development of the Internet of Things and sensor technology, battery temperature management has gradually evolved from single sensor monitoring to multi-sensor fusion solutions, and combined with intelligent algorithms to achieve more accurate temperature control management. By arranging multiple temperature sensors outside and inside the battery, more comprehensive temperature data can be obtained. This multi-point monitoring technology can more accurately describe the temperature distribution of the battery and optimize the measurement results through data fusion technology.

[0027] The purpose of this application is to control the degree of battery heating while accurately judging the internal temperature of the battery, so that the safety test can reflect the internal state of the battery, which helps to improve the consistency of the battery's physical and chemical design and module design.

[0028] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0029] In an exemplary embodiment, Figure 1 As shown, a method for predicting and analyzing thermal runaway of a lithium-ion battery is provided. The method is executed by a computer device, and can be specifically executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In an embodiment of the present application, the method is applied to a server as an example for explanation, and includes the following steps.

[0030] Step 100: Acquire surface temperature data of the lithium-ion battery.

[0031] Step 200: Using a battery thermal simulation model, the estimated battery internal temperature is determined based on the surface temperature data. The battery thermal simulation model is a physical simulation model that is determined by triggering a battery thermal runaway test and based on signal data collected by temperature sensors set outside the lithium-ion battery packaging and at the center of the battery internal winding core surface to characterize the mapping relationship between the internal and external temperature signals of the battery.

[0032] Among them, battery thermal runaway tests include: puncture, overheating and overcharging.

[0033] Step 300: Using a temperature prediction model, predict the internal temperature of the battery according to the surface temperature data. The temperature prediction model is determined using a machine learning method.

[0034] Step 400: Perform fusion processing based on the estimated battery internal temperature and the predicted battery internal temperature to obtain a battery internal temperature estimation result. The battery internal temperature estimation result is used to characterize the thermal state of the lithium-ion battery.

[0035] The battery thermal simulation model includes: heat generation model and heat transfer model.

[0036] Among them, the heat generation model is determined based on the ohmic heat generated by the battery resistance effect, the polarization heat generated when the electrode potential deviates from the equilibrium potential during the charging and discharging process of the battery, and the reaction heat generated by the electrochemical reaction between the electrode material and the electrolyte during the charging and discharging process.

[0037] The heat transfer model is determined based on the heat generated by conduction, convection, radiation and electrochemical reactions of lithium-ion batteries.

[0038] The calculation formula of the heat generation model is: .

[0039] .

[0040] .

[0041] .

[0042] in, is the mathematical expression of the heat generation model; for ohmic heat; is the heat of reaction; is polarization heat; is the working current; is the internal resistance of the battery; is the electrochemical reaction equivalent number of the battery; is the Faraday constant; is the open circuit voltage; is temperature; is the partial derivative of the open circuit voltage with respect to temperature; is the voltage drop caused by the resistor; is the voltage drop caused by the difference in ion concentration; is the voltage drop caused by the electrode activation energy.

[0043] As an optional implementation manner, the method for determining the temperature prediction model specifically includes: Acquire training data; the training data includes surface measurement temperature data of the lithium-ion battery and corresponding label data; the label data includes: internal measurement temperature data.

[0044] Construct an initial decision tree; input the training data into the initial decision tree, and iteratively train and guide the initial decision tree based on the gradient and the second-order gradient to obtain a decision tree; wherein the gradient is the first-order derivative of the loss function, and the second-order gradient is the second-order derivative of the loss function; the loss function is determined based on the label data and the data output by the decision tree.

[0045] The decision tree is weighted and processed based on the regularization term to obtain the temperature prediction model.

[0046] Among them, for any iteration: A split point is selected based on a gain function to determine a split node; a tree structure is generated according to the split node, and a pruning judgment is performed based on an evaluation condition to obtain a first judgment result; the evaluation condition is that the gain of the split node is less than a preset threshold.

[0047] If the first judgment result is yes, the tree structure is pruned and the process of "selecting split points based on the gain function and determining split nodes" is returned.

[0048] If the first judgment result is no, the tree structure is scaled according to the learning rate, and it is determined whether the stopping condition is met to obtain a second judgment result; the stopping condition is that the loss function is within a preset range or the number of iterations reaches the set maximum rounds.

[0049] If the second judgment result is yes, the tree structure corresponding to the current number of iterations is used as the decision tree.

[0050] If the second judgment result is no, the process returns to “selecting split points based on the gain function and determining split nodes”.

[0051] In one embodiment, the calculation formula of the battery internal temperature estimation result is: .

[0052] in, is the estimated result of the internal temperature of the battery; is the first weight; To estimate the internal temperature of the battery; is the second weight; To predict the internal temperature of the battery.

[0053] In practical applications, such as Figure 2 The following is a conceptual flow chart of the method mentioned in this application. The specific operation steps of the method mentioned in this application include: S1: A temperature sensor is arranged outside the lithium-ion battery packaging method and at the center of the surface of the internal winding core of the battery, and a temperature signal device is arranged.

[0054] S2: Trigger thermal runaway of the battery under test by means of needle puncture, overheating, overcharging, etc., and record the temperature signal of the temperature sensor set during the test and other test signals.

[0055] S3: Obtain the mapping relationship between the collected internal and external temperature signals of the battery through a thermal simulation model or machine learning method, and estimate the internal temperature of the battery through the external temperature signal and other test data. Obtain the internal temperature estimation result of the battery through the model fusion formula.

[0056] S4: Repeat the above test under different environments and operating conditions to verify and optimize the battery internal temperature estimation model.

[0057] The method for setting the battery internal core temperature sensor is as follows: According to the requirements of the battery built-in sensor for accurate measurement, corrosion resistance and no impact on the battery structure, the fiber Bragg grating sensor is selected to measure the internal temperature of the battery. At the same time, in order to ensure the stability of the optical fiber in the harsh battery working environment (such as high temperature, corrosion, electrochemical reaction, etc.), a protective layer (such as polyimide or metal coating) is added to the optical fiber.

[0058] The optical fiber is evenly laid on the surface of the battery core, focusing on the positive and negative electrodes and the diaphragm to monitor the local temperature changes of the battery. During installation, the mechanical fixation of the optical fiber sensor must be ensured without affecting the battery structure or performance.

[0059] Lead out the signal lead of the fiber optic sensor from the battery package and connect it to the external data acquisition system. The fiber optic sensor senses the temperature through the change of the optical signal (such as wavelength, intensity, etc.). The acquisition system monitors the change of the optical signal of the fiber optic sensor in real time through a spectrum analyzer or photoelectric converter. The battery sealing needs to be ensured during the packaging process after the lead out to ensure that it does not affect the safety and performance of the battery.

[0060] Thermal simulation model for mapping internal and external temperature signals of the battery, including heat generation model and heat transfer model.

[0061] The calculation formula of the heat generation model of the battery is: .

[0062] Ohmic heat is the heat generated by the resistance effect of the battery. The electrodes, electrolyte, current collector, etc. inside the battery have a certain resistance. When current passes through, according to Joule's law, the resistance inside the battery will generate heat. The calculation formula is: .

[0063] in, is the operating current, is the internal resistance of the battery. As the current increases or the battery ages, the generation of ohmic heat increases significantly.

[0064] During the charging and discharging process of the battery, the electrode material and the electrolyte undergo electrochemical reactions, which are accompanied by heat release or heat absorption. According to the laws of electrochemical thermodynamics, the reaction heat calculation formula is: .

[0065] in, is the electrochemical reaction equivalent number of the battery, is Faraday's constant, is the open circuit voltage, is the partial derivative of the open circuit voltage with respect to temperature.

[0066] Polarization heat is caused by the non-equilibrium state in the electrochemical reaction of the electrode. It is specifically manifested as the heat generated when the electrode potential deviates from the equilibrium potential during the battery's charge and discharge process. Polarization can be divided into resistance polarization, concentration polarization, and activation polarization. The formula for polarization heat is: .

[0067] in, is the voltage drop caused by the resistor, is the voltage drop caused by the difference in ion concentration, is the voltage drop caused by the electrode activation energy.

[0068] The heat transfer process inside and outside the battery involves multiple mechanisms, including conduction, convection, radiation, and heat generated by electrochemical reactions.

[0069] Heat conduction inside the battery mainly occurs between the electrodes, electrolyte, diaphragm, current collector and other components. These components have different thermal conductivity properties, which affects the temperature distribution inside the battery. The rate of heat conduction depends on the thermal conductivity of the material. , the heat conduction equation is Fourier's law: .

[0070] in, is the heat conduction flow, is the temperature gradient.

[0071] When the battery coolant temperature increases due to heat transfer, natural convection occurs. The heat transfer rate of natural convection depends on the temperature difference and the convection coefficient of the fluid. , the heat transfer equation is: .

[0072] in, is the surface area, and are the battery surface and ambient temperature respectively. is the convective heat transfer rate.

[0073] The battery surface radiates heat to the external environment through electromagnetic waves, and the radiation effect is more significant at high temperatures. The calculation of radiant heat is based on the Stefan-Boltzmann law: .

[0074] in, is the emissivity, is the Stefan-Boltzmann constant. is the radiation heat flux.

[0075] Machine learning for mapping the internal and external temperature signals of the battery, specifically: The XGBoost model is used to fit the external and internal temperature data of the battery.

[0076] The battery external temperature data is used as the feature matrix , taking the temperature data measured inside the battery as the label vector .

[0077] Initially, all predictions from the model are a constant, usually the mean of the training data (for regression) or the log odds (for classification). This value is called the initial prediction. .

[0078] XGBoost gradually improves the model through multiple boosting iterations. In each iteration, a new decision tree is constructed to optimize the loss function. Assume that The steps for each round are as follows: Calculate the residual at In the round iteration, the gradient and second-order gradient (i.e. loss function) of each sample are first calculated. The first and second derivatives of ) are used to guide the generation of new trees.

[0079] gradient : For each sample , calculate the error between the current prediction and the true target. It is The label vector of the samples. It is The predicted value corresponding to the round iteration.

[0080] .

[0081] Second Derivative: Calculate the second derivative of the loss function .

[0082] .

[0083] Gradient and second-order derivative information are used to guide the construction of each tree and help improve the stability and accuracy of the model.

[0084] Based on the calculated gradient and second-order derivative information, XGBoost builds a decision tree in each iteration. The decision tree construction process is as follows: Candidate split points: For each feature, several candidate split points are calculated and the contribution of each split point to the gradient and loss function is evaluated.

[0085] Select the optimal split point: Select the split point that can maximize the gain of the objective function. The formula is as follows: .

[0086] in, and are the gradient sums of the left and right child nodes respectively; and is the sum of the second-order derivatives of the left and right child nodes; is the regularization parameter; It is the penalty term for leaf node splitting to prevent overfitting.

[0087] Spanning tree structure: Continuously recursively split nodes until a termination condition is met (such as the maximum depth of the tree or the minimum number of leaf nodes).

[0088] When the decision tree is fully grown, XGBoost checks whether post-pruning is needed. By evaluating the contribution of each leaf node, if the gain of a leaf node is less than a preset threshold (controlled by hyperparameters), it can be pruned to simplify the model and reduce overfitting.

[0089] The newly constructed decision tree will be based on the learning rate Scale it up and then add it to the existing model.

[0090] .

[0091] in, It is The decision tree generated by rounds of iterative training, is the learning rate, which is used to control the contribution of each tree to the model and prevent overfitting. It is The predicted value corresponding to the round iteration.

[0092] After each round of iteration, the loss value of the model is calculated to determine whether the model is converging. The choice of loss function depends on the specific task (such as mean square error MSE, logarithmic loss, cross entropy loss, etc.). The model will continue to iterate until the set maximum number of rounds T1 is reached or the loss no longer decreases significantly.

[0093] After all iterations, XGBoost will get a set of decision trees. The final prediction value is the weighted sum of all trees: .

[0094] In order to control the complexity of the model, XGBoost introduces a regularization term in the objective function, which mainly includes the following two aspects: Leaf node weight regularization: By applying L1 or L2 regularization to the weight of each leaf node, the weight of the leaf node is prevented from being too large to prevent overfitting.

[0095] Tree complexity regularization: By imposing a penalty on the structure of the tree (such as the number of leaf nodes), the model will not be too complex.

[0096] Estimated battery internal temperature for battery thermal simulation models And the temperature prediction model predicts the internal temperature of the battery The final estimated temperature, i.e. the estimated result of the internal temperature of the battery, is obtained by the following formula: .

[0097] .

[0098] in, and are the weights of the battery thermal simulation model and the temperature prediction model, namely is the first weight, is the second weight, which corresponds to the ratio of the temperature estimation accuracy of the two models.

[0099] This application estimates the temperature inside the battery and optimizes the model by setting temperature sensors at the center of the outer and inner core surfaces of the lithium-ion battery package, and combining thermal simulation models or machine learning methods. This has many technical advantages: 1. Improve battery safety: Placing the temperature sensor at the center of the battery's internal winding core allows for more accurate monitoring of temperature changes in the most critical area of ​​the battery. This location is most prone to overheating and thermal runaway, so potential safety hazards can be discovered earlier. Through thermal simulation models or machine learning methods, the external temperature signal is associated with the internal temperature. Even if the battery cannot be directly monitored, the internal temperature can be inferred through external sensors, avoiding the limitations of relying solely on a single external monitoring method.

[0100] 2. Multiple tests and trigger mechanisms: Trigger thermal runaway of the battery through needle puncture, overheating, overcharging, etc., simulating the extreme environment of the battery under real working conditions. This not only ensures the reliability of the sensor, but also obtains comprehensive data under different failure modes to improve the safety performance evaluation of the battery. Different working condition settings can simulate various scenarios of the battery in the real use environment and improve the adaptability and generalization ability of the battery model. This allows the model to more effectively respond to different temperature changes and prevent the occurrence of thermal runaway.

[0101] 3. High-precision temperature estimation: Use thermal simulation models or machine learning methods to establish an accurate mapping relationship between the external and internal temperatures of the battery. These technologies can more accurately estimate the internal temperature of the battery, which not only improves monitoring accuracy but also reduces the need to rely on high-cost internal sensors. Repeated testing in different environments and working conditions, continuously updating and optimizing the temperature estimation model, so that the model can adapt to a variety of usage scenarios and maintain high accuracy under different conditions.

[0102] 4. Accurate modeling and data-driven approach: The machine learning model can autonomously learn according to different environments and working conditions, and optimize the relationship mapping between external sensor signals and internal temperature, improving the robustness and accuracy of the model under complex conditions. The thermal simulation model provides a physics-based theoretical framework, which can be combined with sensor data to conduct deeper temperature analysis and enhance the controllability and predictive ability of the battery management system.

[0103] 5. Reduced cost and complexity: Although temperature sensors are installed internally, the focus is on estimating the internal temperature through external signals, which reduces the need for a large number of internal sensors and reduces system complexity and cost. The arrangement of internal sensors in complex structures is difficult and risky, but this solution effectively avoids this problem by utilizing external data while ensuring monitoring accuracy.

[0104] 6. Support model expansion and customization: This method can be adjusted according to different types of batteries and is applicable to various lithium-ion battery structures, regardless of different packaging forms or different chemical systems. Model optimization and sensor layout can be flexibly adjusted according to specific application scenarios. With the advancement of sensor technology and machine learning algorithms, the solution can be continuously updated to keep pace with the technological frontier and ensure the advancement of the battery temperature management system.

[0105] Based on the same inventive concept, the embodiment of the present application also provides a lithium-ion battery thermal runaway prediction and analysis device for implementing the above-mentioned lithium-ion battery thermal runaway prediction and analysis method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more lithium-ion battery thermal runaway prediction and analysis device embodiments provided below can refer to the limitations of the lithium-ion battery thermal runaway prediction and analysis method above, and will not be repeated here.

[0106] In an exemplary embodiment, a lithium-ion battery thermal runaway prediction and analysis device is provided, comprising: The data acquisition module is used to acquire the surface temperature data of the lithium-ion battery.

[0107] The temperature estimation module is used to use a battery thermal simulation model to determine the estimated internal temperature of the battery based on the surface temperature data; the battery thermal simulation model is a physical simulation model that is determined by triggering a battery thermal runaway test and is used to characterize the mapping relationship between the internal and external temperature signals of the battery based on signal data collected by temperature sensors set outside the lithium-ion battery packaging method and at the center of the surface of the battery internal winding core.

[0108] The temperature prediction module is used to use a temperature prediction model to determine and predict the internal temperature of the battery based on surface temperature data; the temperature prediction model is determined using a machine learning method.

[0109] The fusion processing module is used to perform fusion processing based on the estimated battery internal temperature and the predicted battery internal temperature to obtain the battery internal temperature estimation result; the battery internal temperature estimation result is used to characterize the thermal state of the lithium-ion battery.

[0110] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 3As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store lithium-ion battery thermal runaway prediction and analysis data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a lithium-ion battery thermal runaway prediction and analysis method is implemented.

[0111] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0112] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0113] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0114] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization of the corresponding device owner. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0115] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAM), magnetoresistive random access memories (MRAM), ferroelectric random access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0116] The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., without limitation.

[0117] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0118] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for predicting and analyzing thermal runaway of a lithium-ion battery, characterized in that: The lithium-ion battery thermal runaway prediction and analysis method comprises: Obtain surface temperature data of lithium-ion batteries; A battery thermal simulation model is used to determine the estimated internal temperature of the battery according to the surface temperature data; the battery thermal simulation model is a physical simulation model for characterizing the mapping relationship between the internal and external temperature signals of the battery, determined by triggering a battery thermal runaway test and based on signal data collected by temperature sensors arranged outside the lithium-ion battery packaging and at the center of the surface of the internal winding core of the battery; Using a temperature prediction model to predict the internal temperature of the battery according to the surface temperature data; the temperature prediction model is determined by a machine learning method; A fusion process is performed based on the estimated battery internal temperature and the predicted battery internal temperature to obtain a battery internal temperature estimation result; the battery internal temperature estimation result is used to characterize the thermal state of the lithium-ion battery.

2. The method for predicting and analyzing thermal runaway of a lithium-ion battery according to claim 1, characterized in that: The battery thermal simulation model includes: a heat generation model and a heat transfer model; The heat generation model is determined based on the ohmic heat generated by the battery resistance effect, the polarization heat generated by the deviation of the electrode potential from the equilibrium potential during the battery charging and discharging process, and the reaction heat generated by the electrochemical reaction between the electrode material and the electrolyte during the charging and discharging process; The heat transfer model is determined based on the heat generated by heat conduction, convection, radiation and electrochemical reaction of the lithium-ion battery.

3. The method for predicting and analyzing thermal runaway of a lithium-ion battery according to claim 2, characterized in that: The calculation formula of the heat generation model is: ； ； ； ； in, is the mathematical expression of the heat generation model; for ohmic heat; is the heat of reaction; is polarization heat; is the working current; is the internal resistance of the battery; is the electrochemical reaction equivalent number of the battery; is the Faraday constant; is the open circuit voltage; is temperature; is the partial derivative of the open circuit voltage with respect to temperature; is the voltage drop caused by the resistor; is the voltage drop caused by the difference in ion concentration; is the voltage drop caused by the electrode activation energy.

4. The method for predicting and analyzing thermal runaway of a lithium-ion battery according to claim 1, characterized in that: The method for determining the temperature prediction model specifically includes: Acquire training data; the training data includes surface measurement temperature data of the lithium-ion battery and corresponding label data; the label data includes: internal measurement temperature data; Construct an initial decision tree; Input the training data into the initial decision tree, and perform iterative training guidance on the initial decision tree based on the gradient and the second-order gradient to obtain a decision tree; wherein the gradient is the first-order derivative of the loss function, and the second-order gradient is the second-order derivative of the loss function; and the loss function is determined based on the label data and the data output by the decision tree; Performing weighted sum processing on the decision tree based on the regularization term to obtain the temperature prediction model; Among them, for any iteration: Select the split point based on the gain function and determine the split node; A tree structure is generated according to the split node, and a pruning judgment is performed based on an evaluation condition to obtain a first judgment result; the evaluation condition is that the gain of the split node is less than a preset threshold; If the first judgment result is yes, the tree structure is pruned, and "selecting split points based on the gain function and determining split nodes" is returned; If the first judgment result is no, the tree structure is scaled according to the learning rate, and it is determined whether a stopping condition is met to obtain a second judgment result; the stopping condition is that the loss function is within a preset range or the number of iterations reaches a set maximum round; If the second judgment result is yes, the tree structure corresponding to the current number of iterations is used as the decision tree; If the second judgment result is no, then return to "select splitting points based on gain function and determine splitting nodes".

5. The method for predicting and analyzing thermal runaway of a lithium-ion battery according to claim 1, characterized in that: The calculation formula of the battery internal temperature estimation result is: ； in, is the estimated result of the internal temperature of the battery; is the first weight; To estimate the internal temperature of the battery; is the second weight; To predict the internal temperature of the battery.

6. The method for predicting and analyzing thermal runaway of a lithium-ion battery according to claim 1, characterized in that: The battery thermal runaway test includes: puncture, overheating and overcharging.

7. A lithium-ion battery thermal runaway prediction and analysis device, characterized in that: The lithium-ion battery thermal runaway prediction and analysis device comprises: A data acquisition module, used to acquire surface temperature data of the lithium-ion battery; a temperature estimation module, for determining an estimated internal temperature of the battery according to the surface temperature data using a battery thermal simulation model; the battery thermal simulation model is a physical simulation model for characterizing a mapping relationship between internal and external temperature signals of the battery, determined by triggering a battery thermal runaway test and based on signal data collected by temperature sensors disposed outside the lithium-ion battery packaging and at the center of the surface of the battery internal winding core; A temperature prediction module, used to use a temperature prediction model to determine and predict the internal temperature of the battery according to the surface temperature data; the temperature prediction model is determined by a machine learning method; The fusion processing module is used to perform fusion processing according to the estimated battery internal temperature and the predicted battery internal temperature to obtain a battery internal temperature estimation result; the battery internal temperature estimation result is used to characterize the thermal state of the lithium-ion battery.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for predicting and analyzing thermal runaway of a lithium-ion battery according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting and analyzing thermal runaway of a lithium-ion battery according to any one of claims 1 to 6 is implemented.

Images