Multidimensional Parameter Evaluation System and Method Based on Detection of Thermal Runaway Risk of Battery Cells

Abstract

The present invention relates to the technical field of battery management, and discloses a multi-dimensional parameter evaluation system and method based on the detection of the risk of thermal runaway of battery cells. By collecting multi-dimensional parameter data of battery cells under different charge and discharge states, a thermal-electrical coupling data set is constructed, and electrochemical impedance spectroscopy data under different aging states is obtained to construct an essence multi-dimensional electrochemical feature space; based on this feature space, a battery cell health evaluation model is trained, the battery cell health index is extracted, and the mapping relationship with the thermal runaway risk level is established; combined with the real-time working condition parameters of the battery cell, by dynamically adjusting the weight coefficients of each feature in the health evaluation model, a working condition adaptive battery cell health index is obtained, and based on this, the thermal runaway risk level is predicted to trigger an early warning mechanism; the present invention can realize the accurate, real-time and dynamic evaluation of the thermal runaway risk of battery cells, significantly improve the accuracy and reliability of the thermal runaway early warning of battery cells, and provide strong technical support for battery safety management.

Description

Technical Field

[0001] The present invention relates to the field of battery management technology, and more specifically, to a multi-dimensional parameter evaluation system and method based on battery cell thermal runaway risk detection. Background Art

[0002] With the widespread application of lithium-ion batteries in electric vehicles, energy storage systems and other fields, their safety issues have become increasingly prominent, especially fires and even explosions caused by battery thermal runaway, which seriously threaten personal safety and property safety. Therefore, developing accurate and reliable thermal runaway risk detection methods and achieving early warning are crucial to ensure the safe operation of lithium-ion batteries. In recent years, domestic and foreign scholars have conducted extensive research on battery thermal runaway risk detection and have made some progress, but there are still some problems that need to be solved.

[0003] Existing research has mainly focused on monitoring and analyzing the external characteristic parameters and internal state parameters of battery thermal runaway. For example, the Chinese patent application with publication number CN117630719A discloses a thermal runaway risk warning system, method, device and medium. The system collects the battery state parameters detected by each battery in the battery pack through the first data acquisition component, and the second data acquisition component collects the thermal runaway element parameters detected by the battery pack. The thermal runaway warning component identifies whether the battery pack has a thermal runaway risk based on the above parameters and issues an alarm when the risk is identified. This method mainly relies on the detection of the external characteristic parameters of the battery (such as voltage, temperature, etc.) and the characteristic gases released when thermal runaway occurs (such as hydrogen, carbon monoxide, etc.). However, the changes in these external characteristic parameters often lag behind the changes in the internal state of the battery cell. When obvious external characteristic changes are detected, the battery may already be in a critical state of thermal runaway, or even thermal runaway has already occurred, making it difficult to achieve effective early warning. The Chinese patent application with publication number CN117103997A proposes a battery thermal runaway risk detection method that takes into account the discharge data of the battery system. The method collects the characteristic data of the battery pack, divides the charge and discharge cycles, extracts the discharge data of a specific cycle, and uses the longitudinal outlier mean algorithm to calculate the thermal runaway risk value of each single battery. The single battery that exceeds the risk threshold is regarded as a suspicious battery, and finally further detection and alarm are performed through the judgment mechanism. Although this method takes into account the discharge data of the battery and quantitatively evaluates the thermal runaway risk, it mainly focuses on the performance of the battery in a specific discharge cycle. It lacks a comprehensive evaluation of the battery under different charge and discharge conditions and different aging conditions, and therefore cannot fully reflect the health status and potential risks of the battery. In addition, neither of the above two methods considers the dynamic characteristics of the battery under actual operating conditions, and the actual operating conditions have an important impact on the health status and thermal runaway risk of the battery.

[0004] The prior art fails to fully integrate multi-dimensional parameters characterizing the internal state of the battery cell. In particular, there is a lack of analysis and utilization of the electrochemical impedance spectroscopy data of the battery cell in different aging states, resulting in the inability to comprehensively and accurately evaluate the health state and thermal runaway risk of the battery cell. It also fails to fully consider the dynamic characteristics of the battery cell under actual operating conditions and lacks a real-time monitoring and adaptive adjustment mechanism for operating condition parameters, leading to the inability to accurately evaluate the thermal runaway risk of the battery cell under actual operating conditions. Especially in complex and variable operating conditions, this limitation is more obvious. Summary of the Invention

[0005] To overcome the above defects of the prior art, the present invention provides a multi-dimensional parameter evaluation system and method based on battery cell thermal runaway risk detection. By constructing an essence multi-dimensional electrochemical feature space and combining with a condition-adaptive battery cell health index, it realizes the accurate, real-time, and dynamic evaluation of the battery cell thermal runaway risk. This method can not only accurately evaluate the health state of the battery cell but also effectively predict the thermal runaway risk under different operating conditions, significantly improving the accuracy and reliability of early warning.

[0006] The multi-dimensional parameter evaluation system and method based on battery cell thermal runaway risk detection proposed by the present invention can be widely applied to various application scenarios using lithium-ion batteries as power or energy storage units, such as electric vehicles, buses, energy storage power stations, uninterruptible power supplies, drones, electric bicycles, and home energy storage units. It is especially suitable for scenarios with extremely high safety requirements, such as large-scale energy storage power stations and electric vehicles.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A multi-dimensional parameter evaluation method based on battery cell thermal runaway risk detection, including:

[0009] Synchronously collect multi-dimensional parameter data of the battery cell in different charge and discharge states to construct a thermal-electric coupling data set; obtain the electrochemical impedance spectroscopy data of the battery cell in different aging states, and based on the thermal-electric coupling data set and the electrochemical impedance spectroscopy data, construct an essence multi-dimensional electrochemical feature space; based on the essence multi-dimensional electrochemical feature space, construct and train a battery cell health evaluation model to extract battery cell health indicators; establish a battery cell thermal runaway risk level, and construct a mapping relationship between the battery cell health indicator and the battery cell thermal runaway risk level;

[0010] Obtain the real-time operating condition parameters of the battery cell, and according to the real-time operating condition parameters of the battery cell and the battery cell health evaluation model, obtain a condition-adaptive battery cell health indicator; according to the condition-adaptive battery cell health indicator and the mapping relationship between the battery cell health indicator and the thermal runaway risk level, obtain the predicted thermal runaway risk level; determine whether the predicted thermal runaway risk level exceeds a preset threshold. If it exceeds the preset threshold, trigger an early warning mechanism.

[0011] Furthermore, the multi-dimensional parameter data includes multi-point temperature data, load current characteristic data, and potential response data;

[0012] Constructing the refined multi-dimensional electrochemical feature space based on the thermo-electric coupling data set and the electrochemical impedance spectroscopy data includes:

[0013] Based on the thermo-electric coupling data set, obtaining a non-linear frequency response characteristic matrix;

[0014] Based on the electrochemical impedance spectroscopy data, obtaining an EIS characteristic matrix;

[0015] Fusing the non-linear frequency response characteristic matrix and the EIS characteristic matrix to construct the refined multi-dimensional electrochemical feature space.

[0016] Furthermore, the obtaining of the non-linear frequency response characteristic matrix based on the thermo-electric coupling data set includes:

[0017] Performing time-frequency analysis on the thermo-electric coupling data set to extract multi-parameter wavelet coefficients; the multi-parameter wavelet coefficients include temperature wavelet coefficients, current wavelet coefficients, and voltage wavelet coefficients;

[0018] Based on the multi-parameter wavelet coefficients, obtaining the non-linear frequency response characteristic matrix.

[0019] Furthermore, the extraction of the multi-parameter wavelet coefficients includes:

[0020] Performing time-frequency analysis on the multi-point temperature data to extract temperature wavelet coefficients;

[0021] Performing time-frequency analysis on the load current characteristic data to extract current wavelet coefficients;

[0022] Performing time-frequency analysis on the potential response data to extract voltage wavelet coefficients.

[0023] Furthermore, the obtaining of the non-linear frequency response characteristic matrix based on the multi-parameter wavelet coefficients includes:

[0024] Based on the temperature wavelet coefficients, calculating the energy distribution of the multi-point temperature data in different frequency bands to obtain a temperature frequency-domain feature vector;

[0025] Based on the current wavelet coefficients, calculating the energy distribution of the load current characteristic data in different frequency bands to obtain a current frequency-domain feature vector;

[0026] Based on the voltage wavelet coefficients, calculating the energy distribution of the potential response data in different frequency bands to obtain a voltage frequency-domain feature vector;

[0027] Concatenating the temperature frequency-domain feature vector, the current frequency-domain feature vector, and the voltage frequency-domain feature vector by columns to form the non-linear frequency response characteristic matrix.

[0028] Further, obtaining the EIS feature matrix based on the electrochemical impedance spectroscopy data includes:

[0029] Performing equivalent circuit fitting on the electrochemical impedance spectroscopy data to obtain equivalent circuit model parameters;

[0030] Extracting key impedance parameters from the equivalent circuit model parameters;

[0031] Constructing the EIS feature matrix from the key impedance parameters.

[0032] Further, fusing the nonlinear frequency response feature matrix and the EIS feature matrix to construct the refined multi-dimensional electrochemical feature space includes:

[0033] Performing data standardization processing on the nonlinear frequency response feature matrix and the EIS feature matrix;

[0034] Concatenating the standardized nonlinear frequency response feature matrix and the EIS feature matrix by columns to form a multi-dimensional electrochemical feature matrix; each row of the multi-dimensional electrochemical feature matrix corresponds to a cell sample, and each column corresponds to an electrochemical feature;

[0035] Performing feature selection on the multi-dimensional electrochemical feature matrix;

[0036] Performing feature extraction on the multi-dimensional electrochemical feature matrix after feature selection to obtain the refined multi-dimensional electrochemical feature space, where the refined multi-dimensional electrochemical feature space contains n1 electrochemical features.

[0037] Further, performing feature selection on the multi-dimensional electrochemical feature matrix includes:

[0038] Calculating the importance score of each electrochemical feature in the multi-dimensional electrochemical feature matrix;

[0039] Constructing an electrochemical feature importance matrix based on the importance score of each electrochemical feature;

[0040] Based on the electrochemical feature importance matrix, using the adaptive threshold method to select the optimal feature subset.

[0041] Further, using the adaptive threshold method to select the optimal feature subset based on the electrochemical feature importance matrix includes:

[0042] Designing an adaptive threshold function for electrochemical feature selection to calculate the adaptive threshold for electrochemical feature selection;

[0043] Traversing the electrochemical feature importance matrix and selecting the electrochemical features greater than the adaptive threshold for electrochemical feature selection into the optimal feature subset.

[0044] Further, constructing and training the battery cell health assessment model based on the refined multi-dimensional electrochemical feature space includes:

[0045] Using the electrochemical features in the refined multi-dimensional electrochemical feature space as sample features and the manually labeled battery cell health indicators as sample labels to construct a training sample set;

[0046] Adopting a weighted linear regression model to construct an initial battery cell health assessment model;

[0047] Using the training sample set to train the initial battery cell health assessment model to obtain the weight coefficients and bias term b of each electrochemical feature;

[0048] Based on the weight coefficients and bias term b of each electrochemical feature, obtain the final battery cell health assessment model.

[0049] Further, obtaining the working condition adaptive battery cell health indicator according to the real-time working condition parameters of the battery cell and the battery cell health assessment model includes:

[0050] Obtain the working condition parameters of the battery cell and construct the mapping relationship between the working condition parameters and the weight adjustment coefficient;

[0051] According to the real-time working condition parameters of the battery cell and the mapping relationship between the working condition parameters and the weight adjustment coefficient, calculate the real-time weight adjustment coefficient;

[0052] According to the real-time weight adjustment coefficient, adjust the weight coefficients of each electrochemical feature in the battery cell health assessment model to obtain the adjusted weight coefficients;

[0053] Use the adjusted weight coefficients to calculate the working condition adaptive battery cell health indicator under the current working condition.

[0054] The multi-dimensional parameter evaluation system for battery cell thermal runaway risk detection, which is used to implement the above-mentioned multi-dimensional parameter evaluation method for battery cell thermal runaway risk detection, the system includes:

[0055] Feature space construction module: used to synchronously collect multi-dimensional parameter data of the battery cell under different charge and discharge states to construct a thermal-electric coupling data set; obtain the electrochemical impedance spectroscopy data of the battery cell under different aging states, and based on the thermal-electric coupling data set and the electrochemical impedance spectroscopy data, construct a refined multi-dimensional electrochemical feature space;

[0056] Mapping relationship establishment module: based on the refined multi-dimensional electrochemical feature space, construct and train the battery cell health assessment model, extract the battery cell health indicators; establish the battery cell thermal runaway risk level, and construct the mapping relationship between the battery cell health indicators and the battery cell thermal runaway risk level;

[0057] Thermal runaway judgment module: Obtain the real-time working condition parameters of the battery cell, and obtain the working condition adaptive battery cell health index according to the real-time working condition parameters of the battery cell and the battery cell health evaluation model; obtain the predicted thermal runaway risk level according to the mapping relationship between the working condition adaptive battery cell health index and the thermal runaway risk level corresponding to the battery cell health index; judge whether the predicted thermal runaway risk level exceeds a preset threshold, and if it exceeds the preset threshold, trigger an early warning mechanism.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0059] The multi-dimensional parameter evaluation method based on battery cell thermal runaway risk detection proposed by the present invention innovatively integrates thermal-electrical coupling data and electrochemical impedance spectroscopy data, constructs an essence multi-dimensional electrochemical feature space, and comprehensively characterizes the aging state and thermal runaway risk of the battery cell from multiple dimensions. This method uses a weighted linear regression model to construct a health evaluation model, and dynamically adjusts the feature weights through working condition parameters, realizing accurate evaluation of the health state of the battery cell under different working conditions. In particular, the working condition adaptive battery cell health index proposed by the present invention can reflect the real health state of the battery cell in the actual operating environment in real time, significantly improving the accuracy and practicality of health evaluation. In addition, the mapping relationship established between the battery cell health index and the thermal runaway risk level in the present invention transforms the abstract health index into an intuitive risk level, providing a clear judgment basis for thermal runaway early warning. In summary, the multi-dimensional parameter evaluation system and method proposed by the present invention break through the limitations of traditional single-parameter evaluation methods, realize comprehensive, accurate, real-time, and dynamic evaluation of the thermal runaway risk of the battery cell, provide a strong technical guarantee for battery safety management, effectively reduce the occurrence probability of battery thermal runaway accidents, and are of great significance for improving the safety of application fields such as electric vehicles and energy storage systems, and will have a profound impact on promoting the healthy development of the new energy industry. Description of the Drawings

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0061] Figure 1 It is the principle flow chart of the multi-dimensional parameter evaluation method based on battery cell thermal runaway risk detection in the present invention;

[0062] Figure 2 It is the method flow chart for performing time-frequency analysis on the thermal-electrical coupling data set and extracting multi-parameter wavelet coefficients in the multi-dimensional parameter evaluation method based on battery cell thermal runaway risk detection of the present invention;

[0063] Figure 3 It is the flowchart of the method for obtaining the non - linear frequency response feature matrix based on multi - parameter wavelet coefficients in the multi - dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells according to the present invention;

[0064] Figure 4 It is the flowchart of the method for constructing the EIS feature matrix in the multi - dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells according to the present invention;

[0065] Figure 5 It is the flowchart of the method for fusing the non - linear frequency response feature matrix and the EIS feature matrix to construct the refined multi - dimensional electrochemical feature space in the multi - dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells according to the present invention;

[0066] Figure 6 It is the flowchart of the method for constructing and training the battery cell health assessment model in the multi - dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells according to the present invention;

[0067] Figure 7 It is the flowchart of the method for obtaining the working - condition adaptive battery cell health index in the multi - dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells according to the present invention;

[0068] Figure 8 It is the flowchart of the method for feature selection of the multi - dimensional electrochemical feature matrix in the multi - dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells according to the present invention;

[0069] Figure 9 It is the functional module diagram of the multi - dimensional parameter evaluation system for detecting the risk of thermal runaway of battery cells according to the present invention. Detailed implementation manners

[0070] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0071] Embodiment 1

[0072] Please refer to Figure 1 As shown, this embodiment provides a multi - dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells, including:

[0073] Step S1000, synchronously collect multi-dimensional parameter data of the battery cell under different charge and discharge states to construct a thermal-electrical coupling data set; obtain the electrochemical impedance spectroscopy data of the battery cell under different aging states, and construct an extracted multi-dimensional electrochemical feature space based on the thermal-electrical coupling data set and the electrochemical impedance spectroscopy data; construct and train a battery cell health assessment model based on the extracted multi-dimensional electrochemical feature space, and extract battery cell health indicators; establish a battery cell thermal runaway risk level, and construct a mapping relationship between the battery cell health indicators and the battery cell thermal runaway risk level.

[0074] Further, step S1000 includes:

[0075] Step S1100, synchronously collect multi-dimensional parameter data of the battery cell under different charge and discharge states to construct a thermal-electrical coupling data set; the multi-dimensional parameter data includes multi-point temperature data, load current characteristic data, and potential response data.

[0076] Specifically, in order to obtain comprehensive temperature distribution information, step S1100 sets multiple temperature sensors for collecting the temperature distribution data on the surface and inside of the battery cell. This temperature distribution data, as multi-point temperature data, records the spatial distribution characteristics of the battery cell temperature and provides a data basis for the subsequent fusion of temperature parameters. At the same time, the present invention also synchronously collects the load current characteristic data and potential response data of the battery cell. The load current characteristic data reflects the current change law of the battery cell during the charge and discharge process and embodies the charge and discharge behavior of the battery cell. The potential response data reflects the voltage change law of the battery cell during the charge and discharge process and embodies the electrochemical characteristics of the battery cell. By synchronously collecting the load current characteristic data and potential response data, the present invention can reveal the complex electrochemical process inside the battery cell and provide important data support for subsequent feature extraction and health assessment. After the data collection is completed, the present invention preprocesses the collected thermal-electrical coupling data set, including removing outliers, performing time synchronization and data alignment operations. Data preprocessing can improve data quality, eliminate abnormal interference, and lay a good data foundation for subsequent feature extraction.

[0077] Through the data collection and preprocessing in step S1100, the present invention constructs a comprehensive and reliable thermal-electrical coupling data set. This data set comprehensively considers the thermal characteristics and electrochemical characteristics of the battery cell and provides a solid data foundation for subsequent multi-dimensional parameter fusion and health assessment. By synchronously collecting multi-point temperature data, load current characteristic data, and potential response data, the present invention can deeply analyze the physical and chemical processes inside the battery cell and accurately grasp the evolution law of the battery cell health state. The introduction of data preprocessing further improves data quality and lays a good foundation for subsequent feature extraction and model construction.

[0078] Step S1200, based on the thermoelectric coupling dataset, obtain the non-linear frequency response feature matrix; acquire the electrochemical impedance spectroscopy data of the battery cell in different aging states, and based on the electrochemical impedance spectroscopy data, obtain the EIS feature matrix;

[0079] Further, step S1200 includes:

[0080] Step S1210, based on the thermoelectric coupling dataset, obtain the non-linear frequency response feature matrix;

[0081] Further, step S1210 includes:

[0082] Step S1211, perform time-frequency analysis on the thermoelectric coupling dataset to extract multi-parameter wavelet coefficients; the multi-parameter wavelet coefficients include temperature wavelet coefficients, current wavelet coefficients, and voltage wavelet coefficients;

[0083] Further, as Figure 2 shown, step S1211 includes:

[0084] Step S12111, perform time-frequency analysis on the multi-point temperature data to extract temperature wavelet coefficients;

[0085] Step S12112, perform time-frequency analysis on the load current characteristic data to extract current wavelet coefficients;

[0086] Step S12113, perform time-frequency analysis on the potential response data to extract voltage wavelet coefficients.

[0087] Specifically, time-frequency analysis is a powerful signal processing tool that can capture both the time-domain characteristics and frequency-domain characteristics of a signal simultaneously. Through time-frequency analysis, the variation law of the signal at different time scales and frequency scales can be revealed, and the key features of the signal can be extracted. Wavelet transform is a typical time-frequency analysis method that decomposes a signal at multiple scales to obtain wavelet coefficients in different frequency bands, reflecting the energy distribution characteristics of the signal at different time-frequency scales.

[0088] In sub-step S12111, time-frequency analysis is performed on the multi-point temperature data to extract temperature wavelet coefficients. The temperature data reflects the thermal characteristics of the battery cell. By performing wavelet transform on the temperature data, temperature wavelet coefficients in different frequency bands can be obtained. These temperature wavelet coefficients reflect the change patterns of the temperature signal on different time-frequency scales, embodying the dynamic characteristics of the battery cell temperature. By analyzing the temperature wavelet coefficients, key features such as the instantaneous change and periodic fluctuation of the temperature signal can be captured, providing important information for subsequent temperature feature fusion. In sub-step S12112, time-frequency analysis is performed on the load current characteristic data to extract current wavelet coefficients. The current data reflects the charge and discharge behavior of the battery cell. By performing wavelet transform on the current data, current wavelet coefficients in different frequency bands can be obtained. These current wavelet coefficients reflect the energy distribution characteristics of the current signal on different time-frequency scales, embodying the dynamic characteristics of the current. By analyzing the current wavelet coefficients, key features such as the instantaneous change and periodic fluctuation of the current signal can be captured, revealing the dynamic law of the current during the charge and discharge process of the battery cell, providing important information for subsequent current feature fusion. In sub-step S12113, time-frequency analysis is performed on the potential response data to extract voltage wavelet coefficients. The potential response data reflects the electrochemical characteristics of the battery cell. By performing wavelet transform on the potential response data, voltage wavelet coefficients in different frequency bands can be obtained. These voltage wavelet coefficients reflect the energy distribution characteristics of the voltage signal on different time-frequency scales, embodying the dynamic characteristics of the voltage. By analyzing the voltage wavelet coefficients, key features such as the instantaneous change and periodic fluctuation of the voltage signal can be captured, revealing the dynamic law of the voltage during the charge and discharge process of the battery cell, providing important information for subsequent voltage feature fusion.

[0089] Through step S1211, the present invention realizes the time-frequency analysis of the thermal-electrical coupling data set, and obtains temperature wavelet coefficients, current wavelet coefficients and voltage wavelet coefficients. These wavelet coefficients comprehensively characterize the thermal and electrochemical characteristics of the battery cell from the time-frequency perspective, providing rich feature information for subsequent multi-parameter feature fusion. Through time-frequency analysis, the present invention can deeply excavate the dynamic laws contained in the thermal-electrical coupling data set, capture the early signs of the evolution of the battery cell health state, and achieve a more accurate and comprehensive health assessment.

[0090] Exemplarily, it is assumed that time-frequency analysis is performed on the thermoelectric coupling data set collected in step S1100. For multi-point temperature data, the db4 wavelet basis function is selected, and 5-layer wavelet decomposition is performed on the data of each temperature sensor to obtain temperature wavelet coefficients in different frequency bands. These temperature wavelet coefficients reflect the change patterns of the temperature signal on different time-frequency scales. For example, the low-frequency coefficients reflect the slow-changing trend of the temperature, and the high-frequency coefficients reflect the rapid fluctuations of the temperature. By analyzing the temperature wavelet coefficients, it is found that obvious high-frequency temperature fluctuations occur at the end of the charging of the battery cell, which may indicate local overheating of the battery cell and pose a safety hazard. For the load current characteristic data, the db4 wavelet basis function is selected, and 5-layer wavelet decomposition is performed on the current data to obtain current wavelet coefficients in different frequency bands. By analyzing the current wavelet coefficients, it is found that obvious high-frequency oscillations occur in the current signal when the battery cell is charged and discharged at a large current, which may indicate serious polarization phenomena inside the battery cell and accelerate the capacity decay. For the potential response data, the sym4 wavelet basis function is selected, and 5-layer wavelet decomposition is performed on the potential response data to obtain voltage wavelet coefficients in different frequency bands. By analyzing the voltage wavelet coefficients, it is found that obvious low-frequency fluctuations occur in the voltage signal during the charging and discharging of the battery cell, which may indicate a significant increase in the internal resistance of the battery cell and a decline in the capacity and power characteristics. Through time-frequency analysis, the degradation law of the battery cell health state is revealed from multiple angles, providing an important decision-making basis for subsequent health assessment.

[0091] The application of time-frequency analysis in the present invention greatly expands the analysis dimension of the thermoelectric coupling data set and excavates the rich information contained in the data. By extracting the temperature wavelet coefficients, current wavelet coefficients, and voltage wavelet coefficients, the present invention realizes the dynamic characterization of the thermal and electrochemical characteristics of the battery cell and captures the early signs of the evolution of the battery cell health state. This multi-parameter feature extraction method based on time-frequency analysis lays a solid foundation for subsequent feature fusion and health assessment, helping to improve the accuracy and reliability of health assessment. At the same time, the wavelet coefficients obtained through time-frequency analysis also provide a new perspective for the research on the mechanism of battery cell performance decay, helping to deeply understand the failure mode and thermal runaway mechanism of the battery cell. In short, the time-frequency analysis method adopted in step S1211 provides strong technical support for realizing the health assessment of the battery cell based on multi-dimensional parameters and thermal runaway warning.

[0092] Step S1212: Based on the multi-parameter wavelet coefficients, obtain a non-linear frequency response feature matrix.

[0093] Further, as Figure 3 shown, step S1212 includes:

[0094] Step S12121: Based on the temperature wavelet coefficients, calculate the energy distribution of the multi-point temperature data in different frequency bands to obtain a temperature frequency domain feature vector;

[0095] Step S12122: Calculate the energy distribution of the load current characteristic data in different frequency bands based on the current wavelet coefficients to obtain the current frequency-domain characteristic vector;

[0096] Step S12123: Calculate the energy distribution of the potential response data in different frequency bands based on the voltage wavelet coefficients to obtain the voltage frequency-domain characteristic vector;

[0097] Step S12124: Concatenate the temperature frequency-domain characteristic vector, the current frequency-domain characteristic vector, and the voltage frequency-domain characteristic vector by columns to form a non-linear frequency response characteristic matrix.

[0098] Specifically, the non-linear frequency response characteristic matrix is a characteristic representation method that comprehensively depicts the non-linear dynamic behavior of the battery cell. It integrates the frequency-domain characteristics of three parameters: temperature, current, and voltage, and comprehensively reflects the thermal and electrochemical characteristics of the battery cell. In sub-step S12121, based on the temperature wavelet coefficients, the energy distribution of multi-point temperature data in different frequency bands is calculated to obtain the temperature frequency-domain characteristic vector. The temperature frequency-domain characteristic vector includes the temperature fundamental frequency energy, temperature second harmonic energy, temperature third harmonic energy, etc., which reflects the non-linear frequency response characteristics of the temperature signal. Among them, the temperature fundamental frequency energy represents the energy magnitude of the temperature signal at the fundamental frequency (i.e., the lowest frequency component of the signal), reflecting the overall change trend of the temperature; the temperature second harmonic energy represents the energy magnitude of the temperature signal at twice the fundamental frequency, reflecting the second-order non-linear effect of the temperature; the temperature third harmonic energy represents the energy magnitude of the temperature signal at three times the fundamental frequency, reflecting the third-order non-linear effect of the temperature. By calculating the temperature frequency-domain characteristic vector, the non-linear frequency response characteristics of the temperature signal can be quantitatively evaluated, and the high-order dynamic law of temperature change can be revealed. In sub-step S12122, based on the current wavelet coefficients, the energy distribution of the load current characteristic data in different frequency bands is calculated to obtain the current frequency-domain characteristic vector. The current frequency-domain characteristic vector includes the current fundamental frequency energy, current second harmonic energy, current third harmonic energy, etc., which reflects the non-linear frequency response characteristics of the current signal. Among them, the current fundamental frequency energy represents the energy magnitude of the current signal at the fundamental frequency, reflecting the overall change trend of the current; the current second harmonic energy represents the energy magnitude of the current signal at twice the fundamental frequency, reflecting the second-order non-linear effect of the current; the current third harmonic energy represents the energy magnitude of the current signal at three times the fundamental frequency, reflecting the third-order non-linear effect of the current. By calculating the current frequency-domain characteristic vector, the non-linear frequency response characteristics of the current signal can be quantitatively evaluated, and the high-order dynamic law of current change can be revealed. In sub-step S12123, based on the voltage wavelet coefficients, the energy distribution of the potential response data in different frequency bands is calculated to obtain the voltage frequency-domain characteristic vector. The voltage frequency-domain characteristic vector includes the voltage fundamental frequency energy, voltage second harmonic energy, voltage third harmonic energy, etc., which reflects the non-linear frequency response characteristics of the voltage signal. Among them, the voltage fundamental frequency energy represents the energy magnitude of the voltage signal at the fundamental frequency, reflecting the overall change trend of the voltage; the voltage second harmonic energy represents the energy magnitude of the voltage signal at twice the fundamental frequency, reflecting the second-order non-linear effect of the voltage; the voltage third harmonic energy represents the energy magnitude of the voltage signal at three times the fundamental frequency, reflecting the third-order non-linear effect of the voltage. By calculating the voltage frequency-domain characteristic vector, the non-linear frequency response characteristics of the voltage signal can be quantitatively evaluated, and the high-order dynamic law of voltage change can be revealed.

[0099] In sub-step S12124, the temperature frequency-domain feature vector, current frequency-domain feature vector, and voltage frequency-domain feature vector are concatenated by column to form a non-linear frequency response feature matrix. The non-linear frequency response feature matrix synthesizes the frequency-domain features of the three parameters of temperature, current, and voltage, comprehensively characterizing the non-linear dynamic behavior of the battery cell. By constructing the non-linear frequency response feature matrix, the frequency-domain features of different physical quantities can be integrated into the same mathematical framework, facilitating subsequent feature fusion and analysis. The non-linear frequency response feature matrix not only reflects the frequency-domain characteristics of a single parameter but also reveals the frequency-domain coupling effect between different parameters, providing a new perspective for in-depth understanding of the thermal-electrical coupling behavior of the battery cell. Through step S1212 and its sub-steps, the present invention constructs a non-linear frequency response feature matrix, realizing the comprehensive characterization of the non-linear dynamic behavior of the battery cell. The non-linear frequency response feature matrix fuses the frequency-domain features of the three parameters of temperature, current, and voltage, revealing the high-order dynamic laws of the thermal and electrochemical characteristics of the battery cell, laying a foundation for subsequent multi-dimensional electrochemical feature fusion. Compared with traditional time-domain features, the non-linear frequency response feature matrix explores the internal laws of the state evolution of the battery cell from the frequency-domain perspective, providing a more comprehensive and fine-grained feature representation. By analyzing the non-linear frequency response feature matrix, early signs of battery cell health degradation can be captured, enabling more accurate and timely health state assessment.

[0100] Exemplarily, it is assumed that data acquisition is performed on a battery cell, obtaining time series data of temperature, current, and voltage. Through step S1211, wavelet transform is performed on these data to obtain temperature wavelet coefficients, current wavelet coefficients, and voltage wavelet coefficients. In sub-step S12121, based on the temperature wavelet coefficients, the energy distribution of the temperature signal in different frequency bands is calculated, obtaining the temperature frequency-domain feature vector. For example, the energy proportion of the temperature signal at the fundamental frequency is 80%, the energy proportion at the second harmonic is 15%, and the energy proportion at the third harmonic is 5%. This indicates that the temperature signal is mainly composed of the fundamental frequency component but there are also certain non-linear effects. In sub-steps S12122 and S12123, similar analyses are performed on the current signal and voltage signal, obtaining the current frequency-domain feature vector and voltage frequency-domain feature vector. Finally, in sub-step S12124, these three frequency-domain feature vectors are concatenated by column to form a non-linear frequency response feature matrix. By analyzing the non-linear frequency response feature matrix, it is found that there is a significant increase in the temperature second harmonic energy at the end of charging of the battery cell, and at the same time, the current third harmonic energy also increases. This phenomenon indicates that inside the battery cell, serious electrochemical polarization may occur at a high SOC state, and the risk of thermal runaway rises sharply. Based on this analysis result, the charging strategy can be adjusted in a timely manner to prevent the battery cell from entering a high-risk state.

[0101] In summary, the non-linear frequency response feature matrix constructed in step S1212 provides new ideas and methods for the assessment of the health state of the battery cell. By fusing the frequency-domain features of three parameters, namely temperature, current, and voltage, the non-linear frequency response feature matrix comprehensively characterizes the non-linear dynamic behavior of the battery cell and reveals the internal law of the evolution of the battery cell state. This feature matrix not only enhances the accuracy and reliability of the health state assessment but also lays a solid data foundation for the early warning of battery cell thermal runaway. In practical applications, the non-linear frequency response feature matrix can be combined with other electrochemical features to construct a more complete battery cell health management system to ensure the safe operation of the battery system.

[0102] Step S1220: Obtain the electrochemical impedance spectroscopy data of the battery cell in different aging states;

[0103] Step S1230: Extract the key impedance parameters from the electrochemical impedance spectroscopy data and construct an EIS feature matrix;

[0104] Furthermore, as Figure 4 shown, step S1230 includes:

[0105] Step S1231: Perform equivalent circuit fitting on the electrochemical impedance spectroscopy data to obtain equivalent circuit model parameters;

[0106] Step S1232: Extract the key impedance parameters from the equivalent circuit model parameters;

[0107] Step S1233: Construct an EIS feature matrix from the key impedance parameters.

[0108] Specifically, electrochemical impedance spectroscopy (EIS) is a powerful electrochemical testing technique that can quantitatively evaluate the electrochemical processes inside the battery cell and reveal the evolution law of the battery cell's health state. In an EIS test, a small sinusoidal alternating current excitation signal is applied to the battery cell, and by measuring the response signal of the battery cell, the impedance values at different frequencies are obtained, and then the electrochemical impedance spectrum of the battery cell is plotted. The electrochemical impedance spectrum usually consists of a high-frequency region, a mid-frequency region, and a low-frequency region, reflecting characteristics such as the ohmic impedance, charge transfer impedance, and diffusion impedance inside the battery cell. To comprehensively evaluate the aging state of the battery cell, the present invention collects EIS data at different life cycle stages of the battery cell. Specifically, at key nodes such as the initial state of the battery cell, after 200 cycles, and after 500 cycles, the battery cell is subjected to an EIS test to obtain the electrochemical impedance spectrum data under different aging states. By tracking the evolution trend of the EIS data, the aging behaviors such as the capacity decay, internal resistance increase, and power decline of the battery cell can be quantitatively evaluated, providing important data support for subsequent health state evaluation. When collecting EIS data, it is necessary to strictly control the test conditions to ensure the consistency and reliability of the data. Factors such as the test temperature, state of charge (SOC), and rest time will all affect the test results of EIS, so it is necessary to conduct the test under the same conditions to eliminate the interference of external factors. In addition, it is also necessary to reasonably set the frequency scan range and the number of scan points, taking into account both the test accuracy and efficiency. The frequency scan range usually covers the order of mHz to kHz, and the number of scan points is generally set to 50 - 100 to obtain sufficient frequency resolution.

[0109] The EIS data obtained through step S1220 lays a foundation for subsequent feature extraction and health state evaluation. The EIS data not only reflects the complex electrochemical processes inside the battery cell but also reveals the evolution law of the battery cell's aging state. By tracking the EIS data at different life cycle stages, the dynamic monitoring of the battery cell's health state can be realized, and potential safety hazards of the battery cell can be detected and warned in a timely manner. Based on the analysis of the EIS data, the aging mechanism of the battery cell can be deeply understood, the battery management strategy can be optimized, and the service life of the battery cell can be extended.

[0110] The EIS feature matrix is a compact and efficient data representation form that can condense the rich information contained in EIS data into several key parameters, facilitating subsequent feature fusion and health state assessment. In step S1231, the electrochemical impedance spectroscopy data is fitted with an equivalent circuit to obtain the equivalent circuit model parameters. The equivalent circuit model is a commonly used EIS data analysis method. By constructing a circuit model corresponding to the battery characteristics, quantitative interpretation of the EIS data is achieved. Commonly used equivalent circuit models include the Randles circuit, Thevenin circuit, etc. Among them, the Randles circuit consists of elements such as ohmic internal resistance, double-layer capacitance, charge transfer resistance, and Warburg impedance, and can well describe the electrochemical characteristics of the electrode / electrolyte interface. The Thevenin circuit consists of elements such as ohmic internal resistance, polarization resistance, and polarization capacitance, and can depict the macroscopic characteristics of the battery. By selecting an appropriate equivalent circuit model and using algorithms such as the nonlinear least squares method to fit the EIS data, the key parameters in the equivalent circuit can be obtained to quantitatively describe the internal characteristics of the battery.

[0111] In step S1232, key impedance parameters are extracted from the equivalent circuit model parameters. Different impedance parameters reflect the characteristics of different electrochemical processes inside the battery cell. For example, the charge transfer resistance reflects the resistance of charge transfer at the electrode / electrolyte interface and is closely related to the electrode reaction kinetics; the double-layer capacitance reflects the polarization characteristics of the electrode / electrolyte interface and is related to the electrode specific surface area and roughness; the Warburg impedance reflects the diffusion characteristics of ions in the electrolyte and is related to the electrolyte concentration and diffusion coefficient. By extracting these key impedance parameters, the electrochemical characteristics inside the battery cell can be quantitatively evaluated, and the degradation mechanism of the battery cell health state can be revealed.

[0112] In step S1233, an EIS feature matrix is constructed from the key impedance parameters. The EIS feature matrix is a multi-dimensional data representation form, where each row corresponds to an EIS test sample and each column corresponds to a key impedance parameter. By arranging the impedance parameters of different samples in the same matrix, data comparison and analysis can be conveniently carried out. To eliminate the influence of dimensional differences, the EIS feature matrix also needs to be standardized to unify the value range of each parameter into the interval [0,1] or [-1,1]. Standardization processing can improve the comparability between different parameters and facilitate subsequent feature selection and fusion.

[0113] Through step S1230, the present invention realizes an efficient conversion from EIS data to an EIS feature matrix. The EIS feature matrix not only simplifies the representation form of EIS data but also condenses the key information in the EIS data. By extracting key parameters such as charge transfer resistance, double-layer capacitance, and Warburg impedance, the EIS feature matrix quantitatively characterizes the complex electrochemical characteristics inside the battery cell, laying a foundation for subsequent multi-parameter fusion and state-of-health assessment. Based on the analysis of the EIS feature matrix, aging behaviors such as battery cell capacity attenuation, power decline, and impedance increase can be quantitatively evaluated, revealing the law of the battery cell's health evolution and providing a reliable basis for fault diagnosis and life prediction.

[0114] Step S1300: Integrate the non-linear frequency response feature matrix and the EIS feature matrix to construct an essence multi-dimensional electrochemical feature space; based on the essence multi-dimensional electrochemical feature space, construct and train a battery cell health assessment model, and extract battery cell health indicators; establish a battery cell thermal runaway risk level, and construct a mapping relationship between the battery cell health indicators and the battery cell thermal runaway risk level.

[0115] Furthermore, step S1300 includes:

[0116] Step S1310: Integrate the non-linear frequency response feature matrix and the EIS feature matrix to construct an essence multi-dimensional electrochemical feature space;

[0117] Furthermore, as Figure 5 shown, step S1310 includes:

[0118] Step S1311: Perform data standardization processing on the non-linear frequency response feature matrix and the EIS feature matrix;

[0119] Step S1312: Concatenate the standardized non-linear frequency response feature matrix and the EIS feature matrix by columns to form a multi-dimensional electrochemical feature matrix; each row of the multi-dimensional electrochemical feature matrix corresponds to a battery cell sample, and each column corresponds to an electrochemical feature;

[0120] Step S1313: Perform feature selection on the multi-dimensional electrochemical feature matrix;

[0121] Step S1314: Perform feature extraction on the multi-dimensional electrochemical feature matrix after feature selection to obtain an essence multi-dimensional electrochemical feature space, and the essence multi-dimensional electrochemical feature space contains n1 electrochemical features.

[0122] Specifically, the refined multi-dimensional electrochemical feature space is a highly condensed data representation form that synthesizes various electrochemical features of the battery cell and comprehensively depicts the health state of the battery cell. By integrating non-linear frequency response features and EIS features, the refined multi-dimensional electrochemical feature space fully explores the internal laws contained in the thermal-electric coupling data and impedance spectrum data, laying a solid data foundation for subsequent health assessment.

[0123] Data standardization is a commonly used data preprocessing method that can unify feature data with different dimensions to the same scale and eliminate the influence of dimensional differences on subsequent analysis. Common data standardization methods include max-min standardization, zero-mean standardization, decimal scaling standardization, etc. Among them, max-min standardization linearly scales the feature data into the interval [0,1], zero-mean standardization scales the feature data into an interval with a mean of 0 and a standard deviation of 1, and decimal scaling standardization scales the feature data into the interval [-1,1]. Through data standardization processing, the comparability between different electrochemical features can be improved, facilitating subsequent feature fusion and analysis. The multi-dimensional electrochemical feature matrix is a high-dimensional data representation form that organically combines different types of electrochemical features and comprehensively depicts the health state of the battery cell. By means of column-wise concatenation, the multi-dimensional electrochemical feature matrix integrates non-linear frequency response features and EIS features into the same matrix. Each row corresponds to a battery cell sample, and each column corresponds to an electrochemical feature. This compact data organization form not only facilitates subsequent feature selection and extraction but also provides convenience for constructing a health assessment model.

[0124] Feature selection is a commonly used dimensionality reduction method. By evaluating the correlation, redundancy, and predictive ability of features, it selects the most representative and discriminative feature subset. Common feature selection methods include filter methods, wrapper methods, embedded methods, etc. Among them, filter methods independently evaluate the importance of each feature based on statistical metrics of features (such as variance, correlation coefficient, etc.). Wrapper methods combine feature selection with learning algorithms and obtain the optimal feature subset through iterative search. Embedded methods embed feature selection into learning algorithms and simultaneously achieve feature selection and model training by optimizing the objective function. Through feature selection, redundant and irrelevant features can be removed, the feature dimension can be reduced, and the efficiency and accuracy of subsequent health assessment can be improved. Feature extraction maps the original high-dimensional feature space to a low-dimensional subspace through a certain transformation, realizing data compression and noise reduction while preserving the internal structure of the data. Common feature extraction methods include principal component analysis (PCA), linear discriminant analysis (LDA), independent component analysis (ICA), etc. Among them, PCA converts the original features into linearly independent principal components through orthogonal transformation and sorts them according to the variance size. The first few principal components usually carry most of the information. LDA finds the optimal projection direction by maximizing the between-class scatter and minimizing the within-class scatter to achieve the maximum separation of data of different classes. ICA finds the independent representation of data by maximizing the statistical independence between components. Through feature extraction, while preserving the essential structure of the data, the data dimension can be greatly reduced, and a refined multi-dimensional electrochemical feature space can be obtained. This feature space usually contains n1 electrochemical features, which retain the information of the original data to the greatest extent and provide high-quality input for subsequent health assessment.

[0125] Through step S1310, the present invention constructs a refined multi-dimensional electrochemical feature space, realizing a comprehensive characterization of the health state of the battery cell. The refined multi-dimensional electrochemical feature space integrates non-linear frequency response features and EIS features, fully excavating the internal laws contained in the thermal-electric coupling data and impedance spectrum data, and providing rich feature information for health assessment. Through data standardization processing, the present invention eliminates the dimensional difference between different electrochemical features and improves the comparability of features. By the method of column-wise splicing, the present invention integrates different types of electrochemical features into the same mathematical framework and constructs a multi-dimensional electrochemical feature matrix. Through feature selection and extraction, the present invention removes redundant and irrelevant features, greatly reduces the data dimension while preserving the internal structure of the data, and obtains a refined multi-dimensional electrochemical feature space. Compared with the original thermal-electric coupling data and impedance spectrum data, the refined multi-dimensional electrochemical feature space has higher information density and stronger characterization ability, laying a solid data foundation for subsequent health assessment.

[0126] For example, assume that the non-linear frequency response feature matrix and the EIS feature matrix of 100 battery cell samples are obtained, and each matrix contains 50 features. In sub-step S1311, maximum-minimum normalization processing is performed on these two matrices respectively to scale all feature data into the interval [0, 1]. In sub-step S1312, the two normalized matrices are concatenated by column to form a 100×100 multi-dimensional electrochemical feature matrix. In sub-step S1313, the filter method is used to perform feature selection on the multi-dimensional electrochemical feature matrix, and 30 of the most representative electrochemical features are selected according to the variance and correlation coefficient of the features. In sub-step S1314, PCA is used to perform feature extraction on the matrix after feature selection, and the 30-dimensional feature space is compressed into a 10-dimensional refined multi-dimensional electrochemical feature space. By analyzing the refined multi-dimensional electrochemical feature space, it is found that the battery cell samples in different health states show obvious clustering characteristics in this space, indicating that this feature space has strong characterization ability for the health state of the battery cell. Based on this discovery, an efficient and accurate health assessment model is constructed to realize the quantitative assessment and prediction of the health state of the battery cell.

[0127] In summary, the refined multi-dimensional electrochemical feature space constructed by step S1310 and its sub-steps opens up a new way for the assessment of the health state of the battery cell. By fusing non-linear frequency response features and EIS features, and performing standardization, selection, and extraction on the feature data, the refined multi-dimensional electrochemical feature space retains the internal laws of battery cell health degradation to the greatest extent, providing high-quality input for the construction of the health assessment model. The proposal of this feature space not only expands the analysis dimension of battery cell characterization data but also provides a new idea for multi-source data fusion. In future research, the fusion of the refined multi-dimensional electrochemical feature space with other physical and chemical features can be further explored to construct a more comprehensive and accurate battery cell health assessment system.

[0128] Step S1320: Based on the refined multi-dimensional electrochemical feature space, construct and train a battery cell health assessment model.

[0129] Furthermore, as Figure 6 shown, step S1320 includes:

[0130] Step S1321: Using the electrochemical features in the refined multi-dimensional electrochemical feature space as sample features and the manually labeled battery cell health index as sample labels, construct a training sample set;

[0131] Step S1322: Using a weighted linear regression model, construct an initial battery cell health assessment model;

[0132] Step S1323: Using the training sample set, train the initial battery cell health assessment model to obtain the weight coefficient and bias term b of each electrochemical feature.

[0133] Step S1324: Based on the weight coefficients and bias term b of each electrochemical feature, obtain the final battery cell health assessment model.

[0134] Specifically, the battery cell health assessment model is a mathematical model for quantitatively evaluating the health state of a battery cell. It realizes the quantitative prediction of the battery cell health state by establishing the mapping relationship between the refined multi-dimensional electrochemical features and the battery cell health indicators. The construction and training of the battery cell health assessment model is a data-driven process, which requires a large amount of experimental data and prior knowledge, and uses intelligent algorithms such as machine learning to extract the internal laws of battery cell health degradation from the data. The training sample set is the basis for model training. It consists of a series of input-output data pairs, reflecting the corresponding relationship between the electrochemical features and the health indicators. Among them, the sample features refer to the electrochemical features in the refined multi-dimensional electrochemical feature space, including the fused non-linear frequency response features and EIS features, which comprehensively characterize the health state of the battery cell; the sample labels refer to the manually labeled battery cell health indicators, such as the capacity retention rate, internal resistance increase rate, etc., which quantitatively reflect the health level of the battery cell. By collecting a certain number of battery cell samples and performing standardized charge and discharge tests and electrochemical characterizations on each sample, a comprehensive and reliable training sample set can be obtained. The quality of the sample set directly determines the training effect and generalization performance of the model. Therefore, a reasonable experimental design and data acquisition scheme need to be adopted to cover battery cell samples under different working conditions, different materials, and different production batches as much as possible to improve the representativeness and diversity of the sample set.

[0135] Weighted linear regression establishes a linear relationship between the independent variable and the dependent variable by minimizing the sum of squared errors. In the present invention, the weighted linear regression model takes the refined multi-dimensional electrochemical features as the independent variable and the battery cell health indicators as the dependent variable, and constructs an initial health assessment model through a linear combination. This model has the advantages of simple form and high computational efficiency, and can quickly give a quantitative estimate of the battery cell health state. It should be noted that since different electrochemical features contribute differently to the health indicators, weight coefficients need to be introduced to adjust the importance of each feature. The magnitude of the weight coefficient reflects the relative importance of the corresponding feature, and can be optimized through empirical settings or data-driven methods. In addition, a bias term also needs to be set to correct the overall bias of the model. By reasonably setting the weight coefficients and the bias term, a preliminary battery cell health assessment model can be obtained, providing a basis for subsequent model optimization.

[0136] Model training is a process of parameter optimization. By minimizing the error between the model's predicted values and the true values, the model parameters are continuously adjusted to enable the model to fit the training data as accurately as possible. In the present invention, a training sample set is used to train and optimize the initial battery cell health assessment model. Specifically, the training sample set is input into the model, the predicted values of the model are calculated, and compared with the sample labels to obtain the prediction error. Then, optimization algorithms such as gradient descent are used to adjust the model parameters based on the prediction error to minimize the error. Through multiple iterations of optimization, a set of optimal weight coefficients and bias terms can be obtained, enabling the model to fit the training data to the greatest extent. These weight coefficients quantify the influence degree of different electrochemical characteristics on the health index. The larger the value, the more important the corresponding characteristic; the bias term characterizes the overall deviation of the model and is used to correct the systematic error of the model. During the model training process, methods such as cross-validation need to be adopted to reasonably divide the training set and the validation set to avoid overfitting problems. Through sufficient training and validation, a battery cell health assessment model with excellent performance can be obtained, laying a foundation for subsequent health prediction and thermal runaway warning.

[0137] Through the model training in step S1323, a set of optimal weight coefficients and bias terms are obtained, which characterize the quantitative relationship between the refined multi-dimensional electrochemical characteristics and the battery cell health index. Substituting these parameters into the initial weighted linear regression model, the final battery cell health assessment model is obtained. This model takes the refined multi-dimensional electrochemical characteristics as the input and calculates the quantitative predicted value of the battery cell health index through a weighted linear combination. The higher the predicted value, the better the health state of the battery cell; the lower the predicted value, the worse the health state of the battery cell, and there is a greater risk of thermal runaway. The performance of the health assessment model directly determines the accuracy and reliability of the prediction results. To comprehensively evaluate the model performance, the model needs to be verified and tested on an independent test sample set. By comparing the error between the model's predicted value and the true value and calculating evaluation indicators such as the mean square error and the absolute error, the prediction accuracy and robustness of the model can be quantitatively evaluated. A health assessment model with excellent performance should be able to give accurate and consistent health prediction results on battery cell samples under different working conditions, different materials, and different production batches, providing a reliable decision-making basis for the battery management system.

[0138] Through step S1320 and its sub-steps, the present invention constructs and trains an efficient and accurate battery cell health assessment model. This model takes refined multi-dimensional electrochemical features as input, fully excavates the health degradation information contained in the electrochemical characterization data, overcomes the limitations of single features, and realizes a comprehensive characterization of the battery cell health state. Through machine learning algorithms such as weighted linear regression, this model establishes a quantitative relationship between electrochemical features and health indicators, and can quickly and accurately predict the health state of the battery cell according to its real-time electrochemical performance, providing reliable data support for thermal runaway warning. Compared with the traditional single-index evaluation method, the health assessment model of the present invention has stronger robustness and adaptability, can be applied to battery cells of different types and under different usage conditions, and greatly improves the efficiency and accuracy of health management.

[0139] For example, assume that the performance data of 100 battery cell samples are collected, and each sample is subjected to standardized charge-discharge tests and EIS characterizations. Through step S1310, the non-linear frequency response features and EIS features are fused to obtain a 10-dimensional refined multi-dimensional electrochemical feature space. At the same time, the health indicators of each sample are also obtained through tests such as capacity and internal resistance, and are manually labeled. In step S1321, the refined multi-dimensional electrochemical features are used as sample features, and the manually labeled health indicators are used as sample labels to construct a training set containing 100 samples. In step S1322, a weighted linear regression model is used to construct an initial health assessment model, where the initial values of the weight coefficients and bias terms are randomly set. In step S1323, the training set is used to train and optimize the initial model, and a set of optimal weight coefficients and bias terms are obtained through algorithms such as the least squares method. For example, it is found that the weight coefficients of the charge transfer resistance and double-layer capacitance are the largest, indicating that they have the most significant impact on the health state of the battery cell. In step S1324, the optimal parameters are substituted into the initial model to obtain the final battery cell health assessment model. To evaluate the performance of this model, another 20 independent battery cell samples are collected, subjected to the same tests and characterizations, and the trained model is used for prediction. The results show that the model can predict the health indicators of the battery cell with an accuracy of 95%, and the mean error is less than 5%. This proves that the model has good prediction performance and practical value and can be applied to the actual battery health management system.

[0140] In summary, step S1320 and its sub-steps construct and train an efficient and accurate battery cell health assessment model, providing key technical support for battery health management and thermal runaway warning. This model makes full use of multi-source electrochemical characterization data and, through intelligent algorithms such as machine learning, discovers the internal laws of battery cell health degradation and realizes the quantitative assessment of the health state. The construction idea and training method of the model can not only be extended to other types of battery systems but also provide new solutions for health assessment problems in other engineering fields.

[0141] In step S1330, establish the thermal runaway risk level of the battery cell and construct the mapping relationship between the battery cell health index and the thermal runaway risk level.

[0142] Furthermore, step S1330 includes:

[0143] In step S1331, obtain the battery cell material characteristics and design parameters, analyze the thermal runaway mechanism of the battery cell, and determine the key factors affecting thermal runaway;

[0144] In step S1332, divide the thermal runaway risk levels;

[0145] In step S1333, based on the key factors affecting thermal runaway and the division of risk levels, establish the mapping relationship between the battery cell health index and the thermal runaway risk level.

[0146] Specifically, the thermal runaway risk level is a qualitative risk assessment indicator that divides the thermal runaway risk of the battery cell into several levels, such as low risk, medium risk, high risk, etc., reflecting the possibility and severity of the battery cell experiencing thermal runaway. By establishing the mapping relationship between the health index and the thermal runaway risk level, the quantitative health assessment result can be converted into qualitative risk warning information, providing an intuitive and operable decision-making basis for the battery management system. The thermal runaway mechanism refers to a series of physical and chemical processes of heat accumulation and temperature runaway that occur in the battery cell under specific conditions, such as the decomposition of the positive electrode material, the combustion of the electrolyte, and the melting of the separator. By analyzing the material characteristics of the battery cell (such as the thermal stability of the positive and negative electrode materials, the flammability of the electrolyte, etc.) and design parameters (such as the energy density of the battery cell, the heat dissipation structure, etc.), the root cause of thermal runaway can be revealed from the mechanism, and the key factors affecting thermal runaway can be determined. These key factors may include operating parameters such as the battery cell temperature, SOC, charge and discharge rate, etc., or may also include health indicators such as the battery cell capacity, internal resistance, etc. The analysis of the thermal runaway mechanism is the theoretical basis for risk level division and provides a scientific basis for subsequent risk judgment and warning.

[0147] Based on the results of the thermal runaway mechanism analysis, combined with the safety margin and actual working conditions of the battery cell, the thermal runaway risk is divided into several levels. Common risk level classification methods include the three-level method (low risk, medium risk, high risk), the five-level method (very low risk, low risk, medium risk, high risk, very high risk), etc. The classification of risk levels needs to balance safety and practicality. It should be detailed enough to reflect the differences in risks and concise enough to be easy to operate. When classifying risk levels, scientific and reasonable classification thresholds need to be set, and the risk scope and judgment criteria for each level should be clarified. The formulation of these thresholds needs to comprehensively consider factors such as the material characteristics, design parameters, and working conditions of the battery cell, as well as external factors such as industry standards and regulatory policies. By reasonably classifying risk levels, it can provide a basis for subsequent risk early warning and decision-making.

[0148] The mapping relationship is a method of converting quantitative to qualitative. It maps continuous health indicators to discrete risk levels, realizing the qualitative description of risks. The key to establishing the mapping relationship is to find the corresponding law between the health indicators and risk levels, that is, at what health threshold, the thermal runaway risk of the battery cell reaches what level. This corresponding law can be obtained through a large amount of experimental data and statistical analysis, or set according to the thermal runaway mechanism and expert experience. For example, according to the safety thresholds of parameters such as temperature and SOC, the health indicators can be divided into several intervals, and each interval corresponds to a risk level; or methods such as fuzzy logic and decision trees can be used to establish a non-linear mapping relationship between the health indicators and risk levels. By reasonably constructing the mapping relationship, the health degradation process of the battery cell can be associated with the evolution process of the thermal runaway risk, realizing the dynamic assessment and early warning of risks.

[0149] Through step S1330, the present invention establishes the thermal runaway risk level of the battery cell and constructs the mapping relationship between the health indicator of the battery cell and the thermal runaway risk level. The classification of the thermal runaway risk level is based on the thermal runaway mechanism analysis, fully considering the influence of factors such as the battery cell material, design, and working conditions, and reasonably setting the risk classification threshold, which can accurately reflect the thermal runaway risk level of the battery cell. The mapping relationship between the health indicator and the risk level quantifies the internal connection between the two and realizes the conversion from quantitative to qualitative of risks. By timely evaluating the health indicator of the battery cell and judging its risk level according to the mapping relationship, the real-time monitoring and early warning of the thermal runaway risk can be realized, providing reliable decision-making information for the battery management system.

[0150] The thermal runaway risk level and health - risk mapping is an innovative risk management method that makes up for the deficiencies of traditional methods and realizes the quantification, dynamicization, and visualization of risk assessment. Traditional risk management methods often rely on single indicators (such as temperature, voltage, etc.), lacking overall and systematic perspectives and being unable to accurately grasp the entire process of the health degradation of the battery cell. The method of the present invention is based on the health index, comprehensively considers the influence of multi - source heterogeneous data, comprehensively depicts the evolution law of the battery cell state, and its risk judgment is more scientific and accurate. In addition, the mapping between the health index and the risk level also makes the risk assessment results more intuitive and operable, facilitating the rapid decision - making and feedback control of the battery management system.

[0151] For example, assume that the thermal runaway mechanism analysis of a certain type of battery cell shows that when the temperature exceeds 150 °C and the SOC is higher than 90%, the decomposition of the positive electrode material and the combustion of the electrolyte are extremely likely to occur, and the thermal runaway risk is extremely high. Based on this, the thermal runaway risk levels can be reasonably divided, such as: low risk (temperature below 100 °C, SOC below 70%), medium risk (temperature within 100 - 150 °C, SOC within 70 - 90%), and high risk (temperature above 150 °C, SOC above 90%). On this basis, a mapping relationship between the health index and the risk level is established, such as: when the health index is above 0.8, it is a low risk; when it is between 0.6 - 0.8, it is a medium risk; when it is below 0.6, it is a high risk. In this way, when the health index of a certain battery cell is evaluated to be 0.75, it can be judged that it is at the medium - risk level and certain preventive measures (such as reducing the charging rate, enhancing heat dissipation, etc.) need to be taken; when the health index drops to 0.55, it indicates that it is already at the high - risk level and the use needs to be stopped immediately to avoid the occurrence of thermal runaway accidents.

[0152] In summary, the thermal runaway risk level and health - risk mapping established in step S1330 is an advanced battery safety management technology. Based on the thermal runaway mechanism analysis and taking the health index as a bridge, it closely associates the health degradation process of the battery cell with the evolution process of the thermal runaway risk, realizing real - time monitoring, quantitative early warning, and dynamic control of risks.

[0153] In step S2000, obtain the real - time operating condition parameters of the battery cell, adaptively adjust the weight coefficients of each electrochemical feature in the battery cell health assessment model according to the real - time operating condition parameters of the battery cell to obtain the operating - condition - adaptive battery cell health index; input the operating - condition - adaptive battery cell health index into the pre - established mapping relationship between the battery cell health index and the thermal runaway risk level to obtain the predicted thermal runaway risk level; judge whether the predicted thermal runaway risk level exceeds the preset threshold, and if it exceeds the preset threshold, trigger the warning mechanism.

[0154] Furthermore, step S2000 includes:

[0155] Step S2100: Obtain the real-time working condition parameters of the battery cell, and adaptively adjust the weight coefficients of each electrochemical feature in the battery cell health assessment model according to the real-time working condition parameters of the battery cell to obtain a working condition adaptive battery cell health index;

[0156] Further, step S2100 includes:

[0157] Step S2110: Collect the real-time working condition parameters of the battery cell, where the real-time working condition parameters of the battery cell include the ambient temperature, charge and discharge rate, and battery cell SOC;

[0158] Step S2120: Combine the real-time working condition parameters of the battery cell and dynamically adjust the weight coefficients of each electrochemical feature in the battery cell health assessment model to obtain a working condition adaptive battery cell health index.

[0159] Further, as Figure 7 shown, step S2120 includes:

[0160] Step S2121: Obtain the working condition parameters of the battery cell and construct a mapping relationship between the working condition parameters and the weight adjustment coefficient;

[0161] Step S2122: Calculate the real-time weight adjustment coefficient according to the real-time working condition parameters of the battery cell and the mapping relationship between the working condition parameters and the weight adjustment coefficient;

[0162] Step S2123: Adjust the weight coefficient of each electrochemical feature in the battery cell health assessment model according to the real-time weight adjustment coefficient to obtain the adjusted weight coefficient;

[0163] Step S2124: Use the adjusted weight coefficient to calculate the working condition adaptive battery cell health index under the current working condition.

[0164] Specifically, real-time operating conditions parameters refer to the dynamic parameters of the battery cell during actual use, which reflect the working environment and state of the battery cell and have an important impact on the health degradation and thermal runaway risk of the battery cell. By collecting these parameters in real time, the real-time state of the battery cell can be grasped, providing important input information for subsequent health assessment and risk warning. Among them, the ambient temperature refers to the temperature of the environment where the battery cell is located, which directly affects the electrochemical reaction rate and thermal equilibrium conditions inside the battery cell. Too high ambient temperature will accelerate the capacity attenuation and impedance increase of the battery cell, increasing the thermal runaway risk; too low ambient temperature will reduce the charge and discharge efficiency and energy utilization rate of the battery cell, affecting the working performance of the battery cell. Therefore, real-time monitoring of the ambient temperature is crucial for evaluating the health status of the battery cell and preventing thermal runaway. Commonly used ambient temperature sensors include thermocouples, thermal resistors, thermistors, etc. By reasonably arranging the sensor positions, the temperature distribution information of the environment where the battery cell is located can be obtained. The charge and discharge rate refers to the ratio of the charge and discharge current of the battery cell to the rated capacity, which reflects the charge and discharge intensity and rate of the battery cell. Too high charge and discharge rate will exacerbate the polarization effect and thermal effect inside the battery cell, resulting in a rapid increase in the battery cell temperature and increasing the thermal runaway risk; too low charge and discharge rate will prolong the charge and discharge time and reduce the usage efficiency of the battery cell. Therefore, real-time monitoring of the charge and discharge rate is very necessary for adjusting the charge and discharge strategy of the battery cell and preventing thermal runaway. The charge and discharge rate can be calculated by measuring the charge and discharge current and the rated capacity of the battery cell, or directly obtained through the battery management system. Too high SOC will increase the redox reaction activity of the battery cell, exacerbate the decomposition of the electrolyte and the release of gas, increasing the thermal runaway risk; too low SOC will deepen the over-discharge of the battery cell, resulting in the structural damage and capacity attenuation of the electrode material. Therefore, real-time monitoring of SOC is crucial for optimizing the charge and discharge range of the battery cell and preventing thermal runaway. SOC can be estimated by methods such as Coulomb counting method, open circuit voltage method, Kalman filtering method, etc., or directly obtained through the battery management system. The real-time operating conditions parameters of the battery cell collected through step S2110 provide important decision-making information for subsequent health assessment and risk warning. On the one hand, these parameters reflect the actual working state of the battery cell, affecting the rate and mode of the battery cell health degradation; on the other hand, they are also closely related to the triggering conditions of the battery cell thermal runaway, affecting the level of the thermal runaway risk. By collecting and analyzing these parameters in real time, the battery cell health assessment model can be dynamically adjusted to improve the adaptability and accuracy of the health assessment; at the same time, the abnormal changes in the thermal runaway risk can be detected in time, triggering the corresponding warning mechanism and safety protection measures.

[0165] The operating condition adaptive battery cell health index is a dynamic evaluation index that takes into account the impact of the actual operating conditions of the battery cell on its health state, and can more accurately and timely reflect the health degradation of the battery cell. By obtaining the operating condition parameters in real time and dynamically adjusting the health evaluation model accordingly, the adaptability and robustness of the health evaluation can be significantly improved, providing reliable decision-making support for subsequent risk warning and safety protection.

[0166] The weight adjustment coefficient is an intermediate variable that quantifies the impact of operating conditions. It reflects the contribution degree of electrochemical characteristics to the health state of the battery cell under different operating conditions. By establishing the mapping relationship between the operating condition parameters and the weight adjustment coefficient, the adjustment law of the weight of electrochemical characteristics caused by the change of operating conditions can be quantitatively described. This mapping relationship can be constructed based on mechanism analysis, expert experience or data-driven methods. For example, through theoretical analysis, the quantitative relationship between parameters such as ambient temperature and charge-discharge rate and processes such as electrochemical reaction rate and mass transfer and heat transfer can be deduced, and then the influence law of their weights on electrochemical characteristics can be determined; it is also possible to fit the non-linear mapping relationship between the operating condition parameters and the weight adjustment coefficient through a large amount of experimental data and machine learning methods such as neural networks and support vector machines. The real-time weight adjustment coefficients are a set of dynamically changing quantitative indicators that reflect the relative importance of the weights of each electrochemical characteristic under the current operating conditions in real time. By substituting the real-time collected operating condition parameters into the pre-constructed mapping relationship, the weight adjustment coefficients corresponding to each operating condition parameter can be quickly calculated. It should be noted that since there may be certain measurement errors and random noises in the operating condition parameters, in order to improve the robustness of the calculation, certain data smoothing and filtering methods such as moving average method and Kalman filtering method can be used to preprocess the original operating condition data to reduce the influence of errors and noises.

[0167] The adjusted weight coefficients are a set of dynamically updated model parameters that comprehensively consider the dual influences of operating conditions and electrochemical characteristics, and can more accurately reflect the variation law of the cell health state. The specific adjustment method can adopt mathematical operations such as weighted average and product scaling, combining the original weight coefficients with the corresponding adjustment coefficients to obtain the new weight coefficients. It should be noted that the adjusted weight coefficients need to be normalized to ensure that the value ranges and proportional relationships of each weight coefficient meet the constraint conditions of the model. The operating condition adaptive cell health index is a comprehensive evaluation index that quantifies the cell health state. It fully considers the influences of operating conditions and electrochemical characteristics and can dynamically reflect the instantaneous state of cell health degradation. The specific calculation method is to substitute the real-time collected electrochemical characteristic data into the cell health evaluation model and use the adjusted weight coefficients for weighted combination to obtain a scalar value as the health index at the current moment. Compared with the traditional static health index, the operating condition adaptive cell health index has higher timeliness and pertinence, and can provide a more dynamic and accurate health assessment result for the battery management system.

[0168] Through step S2120, the present invention realizes the operating condition adaptation adjustment of the cell health evaluation model and obtains a new operating condition adaptive cell health index. This index makes full use of the real-time operating condition information of the cell, dynamically adjusts the health evaluation model by quantifying the relationship between the operating condition parameters and the electrochemical characteristic weights, and realizes the real-time update and adaptive optimization of the health state evaluation. Compared with the traditional static health evaluation method, the operating condition adaptive cell health index can more accurately and timely capture the dynamic changes of the cell health state, providing a more reliable and effective basis for the regulation and optimization of the battery management system.

[0169] The proposal of the operating condition adaptive cell health index also provides new research ideas and directions for the field of battery health management. Traditional battery health assessment methods often rely on offline and static models and data, and it is difficult to adapt to complex and changeable operating conditions, and the assessment results are also difficult to reflect the real-time dynamics of the cell health state. However, the operating condition adaptive health assessment method proposed by the present invention realizes the real-time synchronization and adaptive optimization of the health assessment model and the cell operating conditions by real-time sensing of the operating condition parameters and dynamic adjustment of the model weights, providing a new technical means for realizing the health monitoring and management of the entire battery life cycle.

[0170] Step S2200: Input the operating condition adaptive cell health index into the mapping relationship between the cell health index and the thermal runaway risk level to obtain the predicted thermal runaway risk level;

[0171] Specifically, the condition-adaptive battery cell health index is obtained in step S2100. It comprehensively considers the influence of the real-time operating parameters of the battery cell, and adaptively adjusts the battery cell health assessment model, which can more accurately reflect the actual health status of the battery cell under the current operating conditions. The mapping relationship between the battery cell health index and the thermal runaway risk level is established in step S1330. It quantitatively describes the influence of health deterioration on thermal runaway risk and provides a quantitative basis for risk warning. By inputting the condition-adaptive battery cell health index into the health-risk mapping relationship, a predicted thermal runaway risk level can be obtained. This predicted risk level reflects the possibility and severity of thermal runaway of the battery cell under the current health state and operating conditions, and is a qualitative risk description indicator. According to the predicted risk level, the safety status of the battery cell can be preliminarily judged, providing a basis for subsequent risk warning and safety protection.

[0172] Step S2200 organically combines the condition-adaptive health assessment with the qualitative risk level warning, realizing real-time monitoring of the safety status of the battery cell and risk warning. Compared with the traditional threshold warning method based on a single indicator, this method can comprehensively consider the influence of the health status of the battery cell and the operating conditions, and give a more comprehensive and accurate risk prediction result. This prediction result not only reflects the degree of health degradation of the battery cell, but also takes into account the influence of the operating conditions, and can be closer to the actual use of the battery cell. Safety protection measures based on this prediction result will also be more targeted and effective, which can minimize the risk of thermal runaway and ensure the safe operation of the battery.

[0173] For example, suppose that in step S2100, the condition-adaptive cell health index of a cell at a certain moment is 0.75. The index is input into the pre-established health-risk mapping relationship, and the predicted thermal runaway risk level of the cell in the current state is obtained as "medium risk". This prediction result shows that although the overall health status of the cell is acceptable, under the current operating conditions (such as high ambient temperature, large charge and discharge rate, etc.), its thermal runaway risk has reached a high level, and certain preventive measures need to be taken, such as reducing the charge and discharge rate, enhancing heat dissipation, etc. Without these preventive measures, the cell continues to be used in a high-risk state, and its thermal runaway risk will continue to rise, which may cause safety accidents such as fire and explosion in severe cases. Therefore, timely warning and intervention are crucial to ensure battery safety.

[0174] In summary, by combining the working condition adaptive cell health index with the mapping of thermal runaway risk levels, step S2200 realizes the real-time prediction and early warning of the thermal runaway risk of the cell. This method makes full use of the health assessment results and risk level classification results obtained in the previous steps, organically unifies the quantitative health assessment and the qualitative risk level early warning, and forms a complete set of cell safety status monitoring and risk prevention and control systems. This system can dynamically reflect the changes in the cell health status and safety risks, provide a reliable decision-making basis for the battery management system, and maximize the safety operation of the battery. At the same time, the health assessment and risk early warning models on which this method is based are both constructed based on data-driven machine learning methods, and have strong adaptability and scalability. By continuously updating and optimizing the training data, the model can be adapted to different types of cells and working conditions, and its prediction performance will also be continuously improved.

[0175] In step S2300, it is judged whether the predicted thermal runaway risk level exceeds a preset threshold; if it exceeds the preset threshold, the early warning mechanism is triggered in a timely manner, and corresponding safety protection measures are taken.

[0176] Specifically, the preset threshold is reasonably set according to factors such as cell material characteristics, design parameters, and safety margins. It delimits the safety boundary of the thermal runaway risk of the cell. When the predicted risk level exceeds this threshold, it means that the cell has entered a high-risk state. If no measures are taken in a timely manner, the occurrence of a thermal runaway accident will be a high-probability event, and the consequences will also be catastrophic. Therefore, triggering the early warning mechanism in a timely manner and taking effective safety protection measures are crucial for avoiding accidents and ensuring battery safety.

[0177] The setting of the preset threshold needs to consider multiple factors. First, the thermal stability and safety performance of the cell material need to be considered. For cells with different material systems, their characteristics such as thermal runaway triggering temperature and heat release rate vary greatly, which directly affects the level of thermal runaway risk. Therefore, specific recommended values need to be given for different cell materials when setting the preset threshold. Second, the design parameters of the cell, such as energy density and heat dissipation structure, need to be considered. The design parameters affect the temperature distribution and heat accumulation inside the cell, and thus affect the thermal runaway risk. The preset threshold also needs to be adjusted accordingly for different design parameters. Third, the safety margin also needs to be considered. The size of the safety margin determines the sensitivity and reliability of the early warning mechanism. If the margin is set too large, there may be too many false alarms; if the margin is set too small, the risk may not be detected in a timely manner. Therefore, a balanced threshold recommendation needs to be given based on a comprehensive assessment of the cell safety performance.

[0178] When the predicted thermal runaway risk level exceeds the preset threshold, the early warning mechanism will be automatically triggered. The warning information will be promptly transmitted to the battery management system through various channels (such as display screens, audible and visual alarms, communication networks, etc.), and then corresponding safety protection measures will be taken according to the level of the warning and its own functional positioning. These measures may include: reducing the charge and discharge current, increasing the power of the cooling fan, preparing fire extinguishing devices, arranging personnel evacuation, etc. Through these measures, the thermal runaway risk of the battery cells can be effectively reduced, the further deterioration of the accident can be avoided, and the safety of personnel and equipment can be protected.

[0179] In summary, step S2300 constructs a complete early warning and prevention and control mechanism for the thermal runaway risk of battery cells by setting reasonable early warning thresholds and promptly taking safety protection measures when the predicted risk exceeds the standard. This mechanism is based on the risk level prediction result of step S2200 and fully considers the influence of factors such as battery cell materials, design parameters, and engineering applications. It can effectively identify and control the thermal runaway risk of battery cells and improve the safety and reliability of the battery.

[0180] Embodiment 2

[0181] Based on Embodiment 1, this embodiment provides a multi-dimensional parameter evaluation method based on the detection of the thermal runaway risk of battery cells, including:

[0182] Step S1313, perform feature selection on the multi-dimensional electrochemical feature matrix;

[0183] Further, as Figure 8 shown, step S1313 includes:

[0184] Step S13131, calculate the importance score of each electrochemical feature in the multi-dimensional electrochemical feature matrix;

[0185] Step S13132, construct an electrochemical feature importance matrix based on the importance score of each electrochemical feature;

[0186] Step S13133, based on the electrochemical feature importance matrix, use the adaptive threshold method to select the optimal feature subset;

[0187] The method of using the adaptive threshold method to select the optimal feature subset based on the electrochemical feature importance matrix includes:

[0188] Design an adaptive threshold function for electrochemical feature selection to calculate the adaptive threshold for electrochemical feature selection;

[0189] Traverse the electrochemical feature importance matrix and select the electrochemical features greater than the adaptive threshold for electrochemical feature selection into the optimal feature subset.

[0190] Specifically, step S1313 performs feature selection on the multi-dimensional electrochemical feature matrix. The purpose is to screen out the most representative and discriminative feature subset from the high-dimensional feature space, reduce feature redundancy, and improve the generalization performance of the subsequent health assessment model. Feature importance is an indicator to measure the contribution of a single feature to the prediction target. Common calculation methods include feature importance based on tree models (such as average impurity reduction) and feature importance based on regularization (such as L1 regularization coefficient), etc. Considering the possible complex non-linear relationships and interaction effects between electrochemical features, the present invention adopts a feature importance measurement method based on the Shapley value. The Shapley value originates from game theory and is used to evaluate the contribution of each participant in the game. Introducing it into the calculation of feature importance can measure the average marginal contribution of each feature in all possible feature combinations, comprehensively considering the interaction effects between features. Specifically, for each training sample, all possible feature subsets are calculated, the prediction performance of each subset is evaluated, and then the Shapley value calculation formula is used to quantify the average marginal contribution of each feature as its importance score. By performing this process on all training samples and taking the average value, a robust feature importance measurement can be obtained. Compared with traditional methods, the Shapley value can reveal the "net" contribution of features, overcome the influence of multicollinearity between features, and make the importance evaluation more accurate and fair. Arranging the Shapley importance scores of each feature according to the layout of the original feature matrix forms an importance matrix with the same shape as the multi-dimensional electrochemical feature matrix. This matrix intuitively depicts the relative contributions of different electrochemical features to health assessment, providing a quantitative basis for subsequent feature screening. By performing visual analysis on the importance matrix, such as drawing importance distribution maps, clustering heat maps, etc., the overall distribution characteristics of feature importance can be insighted, possible feature clusters and outliers can be found, and prior knowledge can be provided for the formulation of feature selection strategies.

[0191] The key to feature selection is to determine the optimal feature combination, which can retain the most original information while minimizing redundancy and noise to the greatest extent. Traditional feature selection methods usually rely on preset fixed thresholds, such as variance thresholds, correlation coefficient thresholds, etc., lacking flexibility and adaptability. Considering the diversity and non-stationarity of the distribution of electrochemical feature importance, the present invention designs an adaptive threshold feature selection method. Specifically, an adaptive threshold function threshold1 = base value + k × range is defined, where base valueThe basis threshold is [basis threshold], range is the value range of the importance score, and k is the adaptive coefficient. By optimizing the value of k, the threshold can be adaptively adjusted to the dynamic distribution of the importance score, avoiding both the selection of too many redundant features and retaining the key information features. After determining the optimal value of k, traverse the electrochemical feature importance matrix and select all features with importance scores greater than the adaptive threshold to form the final feature subset. This method can automatically adjust the granularity of feature selection from a data-driven perspective, overcoming the subjectivity and limitations of manually setting thresholds and making the selection results more robust and reliable.

[0192] Through the above steps, the automatic screening from the high-dimensional electrochemical feature matrix to the low-dimensional optimal feature subset is realized, effectively reducing the redundancy of the feature space and highlighting the key information that is most decisive for the health assessment. Compared with traditional feature selection methods, the innovations of the present invention are reflected in: (1) Using the Shapley value to measure feature importance, overcoming the influence of multicollinearity among features and obtaining a more accurate and fair importance assessment result; (2) Constructing an electrochemical feature importance matrix to intuitively depict the characteristics of the importance distribution and providing prior knowledge for threshold selection; (3) Introducing the adaptive threshold method to automatically adjust the feature selection granularity according to the dynamic characteristics of the importance distribution and obtaining a more robust and reliable feature subset. These innovations are organically combined to form a set of data-driven electrochemical feature selection methods, laying a high-quality data foundation for the construction of subsequent health assessment models.

[0193] Embodiment 3

[0194] Based on Embodiment 2, this embodiment provides a multi-dimensional parameter evaluation method for detecting the risk of thermal runaway of battery cells, including:

[0195] Step S1313, perform feature selection on the multi-dimensional electrochemical feature matrix;

[0196] Further, step S1313 includes:

[0197] Step S13131, calculate the importance score of each electrochemical feature in the multi-dimensional electrochemical feature matrix;

[0198] Step S13132, construct an electrochemical feature importance matrix based on the importance score of each electrochemical feature;

[0199] Step S13133, select the optimal feature subset using the adaptive threshold method based on the electrochemical feature importance matrix;

[0200] The selection of the optimal feature subset using the adaptive threshold method based on the electrochemical feature importance matrix includes:

[0201] Design an adaptive threshold function for electrochemical feature selection and calculate the adaptive threshold for electrochemical feature selection;

[0202] Traverse the electrochemical feature importance matrix and select the electrochemical features greater than the adaptive threshold for electrochemical feature selection into the optimal feature subset.

[0203] The adaptive threshold function for electrochemical feature selection includes:

[0204] Calculate the skewness coefficient based on the importance scores of electrochemical features;

[0205] Calculate the interquartile range of all electrochemical feature importance scores , according to , obtain the dispersion coefficient;

[0206] Obtain the number of columns of the electrochemical feature importance matrix and calculate the logarithm of the number of features according to the number of columns of the electrochemical feature importance matrix;

[0207] The adaptive threshold for electrochemical feature selection is obtained by weighted combination of the skewness coefficient, dispersion coefficient, and logarithm of the number of features.

[0208] ;

[0209] Among them:

[0210] Threshold: The adaptive threshold for electrochemical feature selection.

[0211] : The importance score of the i-th electrochemical feature.

[0212] : The mean of all electrochemical feature importance scores.

[0213] n: The total number of electrochemical features, that is, the number of columns of the electrochemical feature importance matrix.

[0214] : The standard deviation of all electrochemical feature importance scores.

[0215] : The interquartile range of all electrochemical feature importance scores, obtained by taking the difference between the first and third quartiles of the electrochemical feature importance matrix.

[0216] : The skewness coefficient, reflecting the asymmetry of the importance score distribution;

[0217] : The weight of the skewness coefficient, controlling the influence of the skewness coefficient on the threshold, The larger the value, the more attention is paid to the asymmetry of the importance score distribution, and it is hoped to more selectively retain important features by increasing the threshold.

[0218] : Coefficient of dispersion, which reflects the dispersion degree of the importance score distribution.

[0219] : Weight of the coefficient of dispersion, which controls the influence of the coefficient of dispersion on the threshold. The larger the value, the more attention is paid to the dispersion degree of the importance score, and it is hoped to more comprehensively retain important features by decreasing the threshold.

[0220] : Logarithm of the number of features, which reflects the complexity of the feature selection task;

[0221] : Weight of the logarithm of the number of features, which controls the influence of the number of features on the threshold. The larger the value, the more attention is paid to the complexity of the feature selection task, and it is hoped to control the size of the selected feature subset by increasing the threshold.

[0222] s: Scale parameter in the coefficient of dispersion, which controls the sensitivity of the influence of the ratio of the interquartile range to the standard deviation on the coefficient of dispersion. The larger the value of s, the less sensitive the coefficient of dispersion is to the change of the ratio of the interquartile range to the standard deviation, and the stronger the adaptability of the threshold function to the dispersion degree of the importance score.

[0223] These four hyperparameters jointly determine the adaptability and selection strategy of the threshold function to the electrochemical feature importance matrix. By adjusting these hyperparameters, the threshold function can achieve optimal feature selection performance under different datasets and task requirements. The hyperparameters can be tuned by methods such as cross-validation.

[0224] This formula measures the asymmetry of the importance score distribution by calculating the skewness coefficient of the importance score (i.e., the third central moment divided by the cube of the standard deviation). The larger the skewness coefficient, the more asymmetric the distribution, and the threshold should be adjusted accordingly higher to more selectively retain important features. The dispersion degree of the importance score is measured by calculating the negative exponent of the ratio of the interquartile range to the standard deviation of the importance score. The greater the dispersion degree, the more obvious the difference between important features and unimportant features. The complexity of the feature selection task is measured by taking the logarithm of the total number of features. The more features there are, the higher the task complexity, and the threshold should be increased accordingly to ensure that the selected features are important enough while controlling the subset size. Through the weighted combination of the above three factors, the adaptive threshold function can dynamically adjust the threshold level according to the actual distribution characteristics of the electrochemical feature importance scores, achieving more flexible and effective feature selection.

[0225] The adaptive threshold function for electrochemical feature selection can automatically adjust the threshold level according to the statistical characteristics of different electrochemical feature importance matrices, without the need for manual setting of a fixed threshold, which improves the adaptive ability of feature selection. By introducing the high-order moment (skewness) and dispersion measure (interquartile range) of the importance score distribution, the threshold function is more robust to outliers and noise, and can still make reliable selection decisions in the case of an unbalanced importance score distribution or the presence of outliers. The design of the threshold function takes into account multiple key features of the importance score distribution, such as skewness, dispersion, the number of features, etc., and these features have clear statistical meanings, making the behavior of the threshold function more transparent and interpretable.

[0226] Regarding the output of the threshold function, as the skewness of the importance score distribution increases, the dispersion decreases, and the number of features increases, the output value of the threshold function will increase accordingly. This means that when the difference between important features and unimportant features is less obvious and the number of features is larger, the threshold function will automatically raise the selection criteria to ensure that the selected feature subset is sufficiently refined and important. Conversely, when the difference between important features and unimportant features is more obvious and the number of features is smaller, the threshold function will appropriately lower the selection criteria to avoid missing valuable features. In summary, this adaptive threshold function for electrochemical feature selection can dynamically adjust the selection threshold according to the actual distribution characteristics of the electrochemical feature importance matrix, adaptively balance the size and importance of the feature subset, improve the flexibility, robustness, and interpretability of feature selection, and provide a higher-quality feature subset for the construction of subsequent battery health assessment models.

[0227] Example 4

[0228] Based on Example 1, this example provides a multi-dimensional parameter evaluation system for detecting the risk of thermal runaway of battery cells, such as Figure 9 shown, including:

[0229] Feature space construction module: used to synchronously collect multi-dimensional parameter data of the battery cell under different charge and discharge states, and construct a thermal-electrical coupling data set; obtain the electrochemical impedance spectroscopy data of the battery cell under different aging states, and construct an essence multi-dimensional electrochemical feature space based on the thermal-electrical coupling data set and the electrochemical impedance spectroscopy data;

[0230] Mapping relationship establishment module: based on the essence multi-dimensional electrochemical feature space, construct and train a battery cell health assessment model, and extract battery cell health indicators; establish a battery cell thermal runaway risk level, and construct a mapping relationship between the battery cell health indicators and the battery cell thermal runaway risk level;

[0231] Thermal runaway judgment module: Obtain the real-time working condition parameters of the battery cell, and according to the real-time working condition parameters of the battery cell and the battery cell health assessment model, obtain the working condition adaptive battery cell health index; According to the mapping relationship between the working condition adaptive battery cell health index and the thermal runaway risk level, obtain the predicted thermal runaway risk level; Judge whether the predicted thermal runaway risk level exceeds the preset threshold, and if it exceeds the preset threshold, trigger the warning mechanism.

[0232] In the feature space construction module, the multi-dimensional parameter data includes multi-point temperature data, load current characteristic data and potential response data; The construction of the refined multi-dimensional electrochemical feature space based on the thermo-electrochemical coupling data set and the electrochemical impedance spectroscopy data includes:

[0233] Step S1210, based on the thermo-electrochemical coupling data set, obtain the non-linear frequency response characteristic matrix;

[0234] Step S1220, obtain the electrochemical impedance spectroscopy data of the battery cell in different aging states;

[0235] Step S1230, extract the key impedance parameters from the electrochemical impedance spectroscopy data and construct the EIS characteristic matrix;

[0236] Step S1310, fuse the non-linear frequency response characteristic matrix and the EIS characteristic matrix to construct the refined multi-dimensional electrochemical feature space.

[0237] The said step S1210 includes:

[0238] Step S1211, perform time-frequency analysis on the thermo-electrochemical coupling data set and extract multi-parameter wavelet coefficients; The multi-parameter wavelet coefficients include temperature wavelet coefficients, current wavelet coefficients and voltage wavelet coefficients;

[0239] Step S1212, based on the multi-parameter wavelet coefficients, obtain the non-linear frequency response characteristic matrix.

[0240] The said step S1211 includes:

[0241] Step S12111, perform time-frequency analysis on the multi-point temperature data and extract the temperature wavelet coefficients;

[0242] Step S12112, perform time-frequency analysis on the load current characteristic data and extract the current wavelet coefficients;

[0243] Step S12113, perform time-frequency analysis on the potential response data and extract the voltage wavelet coefficients.

[0244] The said step S1212 includes:

[0245] Step S12121: Calculate the energy distribution of multi-point temperature data in different frequency bands based on the temperature wavelet coefficients to obtain the temperature frequency-domain feature vector;

[0246] Step S12122: Calculate the energy distribution of the load current characteristic data in different frequency bands based on the current wavelet coefficients to obtain the current frequency-domain feature vector;

[0247] Step S12123: Calculate the energy distribution of the potential response data in different frequency bands based on the voltage wavelet coefficients to obtain the voltage frequency-domain feature vector;

[0248] Step S12124: Concatenate the temperature frequency-domain feature vector, the current frequency-domain feature vector, and the voltage frequency-domain feature vector by columns to form a non-linear frequency response feature matrix.

[0249] The said step S1310 includes:

[0250] Step S1311: Perform data standardization processing on the non-linear frequency response feature matrix and the EIS feature matrix;

[0251] Step S1312: Concatenate the standardized non-linear frequency response feature matrix and the EIS feature matrix by columns to form a multi-dimensional electrochemical feature matrix; each row of the multi-dimensional electrochemical feature matrix corresponds to a cell sample, and each column corresponds to an electrochemical feature;

[0252] Step S1313: Perform feature selection on the multi-dimensional electrochemical feature matrix;

[0253] Step S1314: Perform feature extraction on the multi-dimensional electrochemical feature matrix after feature selection to obtain a refined multi-dimensional electrochemical feature space, and the refined multi-dimensional electrochemical feature space contains n1 electrochemical features.

[0254] In the mapping relationship establishment module, the construction and training of the cell health assessment model include:

[0255] Step S1321: Construct a training sample set with the electrochemical features in the refined multi-dimensional electrochemical feature space as sample features and the manually labeled cell health indicators as sample labels;

[0256] Step S1322: Use a weighted linear regression model to construct an initial cell health assessment model;

[0257] Step S1323: Use the training sample set to train the initial cell health assessment model to obtain the weight coefficients and bias term b of each electrochemical feature;

[0258] Step S1324: Obtain the final cell health assessment model based on the weight coefficients and bias term b of each electrochemical feature.

[0259] In the mapping relationship establishment module, the establishment of the mapping relationship between the battery cell health index and the battery cell thermal runaway risk level includes:

[0260] Step S1331: Obtain the battery cell material characteristics and design parameters, analyze the thermal runaway mechanism of the battery cell, and determine the key factors affecting thermal runaway;

[0261] Step S1332: Divide the thermal runaway risk level;

[0262] Step S1333: Based on the key factors affecting thermal runaway and the risk level division, establish the mapping relationship between the battery cell health index and the thermal runaway risk level.

[0263] In the thermal runaway judgment module, obtaining the operating condition adaptive battery cell health index according to the real-time operating condition parameters of the battery cell and the battery cell health assessment model includes:

[0264] Step S2110: Collect the real-time operating condition parameters of the battery cell, where the real-time operating condition parameters of the battery cell include the ambient temperature, charge-discharge rate, and battery cell SOC;

[0265] Step S2120: Combine the real-time operating condition parameters of the battery cell, dynamically adjust the weight coefficients of each electrochemical feature in the battery cell health assessment model, and obtain the operating condition adaptive battery cell health index.

[0266] The said Step S2120 includes:

[0267] Step S2121: Obtain the operating condition parameters of the battery cell, and construct the mapping relationship between the operating condition parameters and the weight adjustment coefficient;

[0268] Step S2122: According to the real-time operating condition parameters of the battery cell and the mapping relationship between the operating condition parameters and the weight adjustment coefficient, calculate the real-time weight adjustment coefficient;

[0269] Step S2123: According to the real-time weight adjustment coefficient, adjust the weight coefficient of each electrochemical feature in the battery cell health assessment model to obtain the adjusted weight coefficient;

[0270] Step S2124: Use the adjusted weight coefficient to calculate the operating condition adaptive battery cell health index under the current operating condition.

[0271] The methods and systems of the present application can be implemented in many ways. For example, the methods and systems of the present application can be implemented through software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of the steps for the method is only for illustration, and the steps of the method of the present application are not limited to the above specifically described order unless otherwise specifically stated.

[0272] In addition, the parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive elaboration.

[0273] As described above in the specific embodiments, the objectives, technical solutions, and beneficial effects of the present invention have been further described in detail. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells, characterized in that: The method comprises: Synchronously collect multi-dimensional parameter data of battery cells under different charge and discharge states to construct a thermal-electric coupling data set; obtain electrochemical impedance spectroscopy data of battery cells under different aging states, and construct a refined multi-dimensional electrochemical feature space based on the thermal-electric coupling data set and electrochemical impedance spectroscopy data; construct and train a battery cell health assessment model based on the refined multi-dimensional electrochemical feature space, and extract battery cell health indicators; establish a battery cell thermal runaway risk level, and construct a mapping relationship between battery cell health indicators and battery cell thermal runaway risk level; Obtain the real-time working condition parameters of the battery cell, and obtain the working condition adaptive battery cell health index according to the real-time working condition parameters of the battery cell and the battery cell health assessment model; obtain the predicted thermal runaway risk level according to the working condition adaptive battery cell health index and the mapping relationship between the battery cell health index and the thermal runaway risk level; determine whether the predicted thermal runaway risk level exceeds a preset threshold, and if so, trigger the early warning mechanism; The method for constructing the mapping relationship between the battery cell health index and the battery cell thermal runaway risk level is: Obtain the material characteristics and design parameters of the battery cell, analyze the thermal runaway mechanism of the battery cell, and determine the key factors affecting thermal runaway; classify the thermal runaway risk level; based on the key factors affecting thermal runaway and the risk level classification, establish a mapping relationship between the battery cell health index and the thermal runaway risk level; The construction of a refined multi-dimensional electrochemical feature space comprises: Based on the thermal-electric coupling data set, a nonlinear frequency response characteristic matrix is ​​obtained; based on the electrochemical impedance spectroscopy data, an EIS characteristic matrix is ​​obtained; data standardization is performed on the nonlinear frequency response characteristic matrix and the EIS characteristic matrix; the standardized nonlinear frequency response characteristic matrix and the EIS characteristic matrix are spliced ​​by columns to form a multidimensional electrochemical characteristic matrix; feature selection is performed on the multidimensional electrochemical characteristic matrix, and feature extraction is performed on the multidimensional electrochemical characteristic matrix after feature selection to obtain a refined multidimensional electrochemical feature space.

2. The multidimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 1, characterized in that: The multi-dimensional parameter data includes multi-point temperature data, load current characteristic data and potential response data.

3. The multidimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 2 is characterized in that: The nonlinear frequency response characteristic matrix obtained based on the thermal-electric coupling data set includes: Performing time-frequency analysis on the thermal-electric coupling data set to extract multi-parameter wavelet coefficients; the multi-parameter wavelet coefficients include temperature wavelet coefficients, current wavelet coefficients, and voltage wavelet coefficients; Based on the multi-parameter wavelet coefficients, the nonlinear frequency response characteristic matrix is ​​obtained.

4. The multidimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 3 is characterized in that: The extracting of multi-parameter wavelet coefficients comprises: Perform time-frequency analysis on multi-point temperature data and extract temperature wavelet coefficients; Perform time-frequency analysis on load current characteristic data and extract current wavelet coefficients; The potential response data is subjected to time-frequency analysis and the voltage wavelet coefficients are extracted.

5. The multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 3 is characterized in that: The method of obtaining a nonlinear frequency response characteristic matrix based on multi-parameter wavelet coefficients includes: Based on the temperature wavelet coefficient, the energy distribution of multi-point temperature data in different frequency bands is calculated to obtain the temperature frequency domain feature vector; Based on the current wavelet coefficient, the energy distribution of the load current characteristic data in different frequency bands is calculated to obtain the current frequency domain characteristic vector; Based on the voltage wavelet coefficient, the energy distribution of the potential response data in different frequency bands is calculated to obtain the voltage frequency domain feature vector; The temperature frequency domain eigenvector, the current frequency domain eigenvector and the voltage frequency domain eigenvector are concatenated column by column to form a nonlinear frequency response characteristic matrix.

6. The multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 1, characterized in that: The EIS characteristic matrix obtained based on the electrochemical impedance spectroscopy data includes: Perform equivalent circuit fitting on electrochemical impedance spectroscopy data to obtain equivalent circuit model parameters; Extract key impedance parameters from equivalent circuit model parameters; The EIS characteristic matrix is ​​constructed from key impedance parameters.

7. The multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 1, characterized in that: Each row of the multidimensional electrochemical feature matrix corresponds to a battery cell sample, and each column corresponds to an electrochemical feature; the refined multidimensional electrochemical feature space contains n1 electrochemical features.

8. The multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 7, characterized in that: The feature selection of the multidimensional electrochemical feature matrix comprises: Calculate the importance score of each electrochemical feature in the multidimensional electrochemical feature matrix; Based on the importance score of each electrochemical feature, an electrochemical feature importance matrix is ​​constructed; Based on the electrochemical feature importance matrix, an adaptive threshold method was used to select the optimal feature subset.

9. The multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 8, characterized in that: The method of selecting the optimal feature subset based on the electrochemical feature importance matrix by using an adaptive threshold method includes: Design an electrochemical feature selection adaptive threshold function and calculate the electrochemical feature selection adaptive threshold; The electrochemical feature importance matrix is ​​traversed, and the electrochemical features greater than the electrochemical feature selection adaptive threshold are selected into the optimal feature subset.

10. The multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 7, characterized in that: The construction and training of the battery cell health assessment model based on the refined multi-dimensional electrochemical feature space includes: The electrochemical features in the refined multi-dimensional electrochemical feature space are used as sample features, and the manually labeled battery cell health indicators are used as sample labels to construct a training sample set; A weighted linear regression model is used to build an initial battery cell health assessment model; Using the training sample set, train the initial cell health assessment model to obtain the weight coefficient and bias term b of each electrochemical feature; Based on the weight coefficient and bias term b of each electrochemical feature, the final cell health assessment model is obtained.

11. The multi-dimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to claim 10, characterized in that: The condition-adaptive battery health index is obtained according to the battery real-time condition parameters and the battery health evaluation model, including: Obtain the operating parameters of the battery cell and construct a mapping relationship between the operating parameters and the weight adjustment coefficient; Calculate the real-time weight adjustment coefficient according to the real-time operating parameters of the battery cell and the mapping relationship between the operating parameters and the weight adjustment coefficient; According to the real-time weight adjustment coefficient, the weight coefficient of each electrochemical feature in the battery cell health assessment model is adjusted to obtain an adjusted weight coefficient; Use the adjusted weight coefficient to calculate the condition-adaptive cell health index under the current condition.

12. A multidimensional parameter evaluation system based on thermal runaway risk detection of battery cells, which is used to implement the multidimensional parameter evaluation method based on thermal runaway risk detection of battery cells according to any one of claims 1 to 11, characterized in that: The system comprises: Feature space construction module: used to synchronously collect multi-dimensional parameter data of battery cells under different charge and discharge states to construct a thermal-electric coupling data set; obtain electrochemical impedance spectroscopy data of battery cells under different aging states, and construct a refined multi-dimensional electrochemical feature space based on the thermal-electric coupling data set and electrochemical impedance spectroscopy data; Mapping relationship establishment module: Based on the refined multi-dimensional electrochemical feature space, build and train the battery cell health assessment model and extract the battery cell health indicators; establish the battery cell thermal runaway risk level and construct the mapping relationship between the battery cell health indicators and the battery cell thermal runaway risk level; Thermal runaway judgment module: obtains the real-time operating parameters of the battery cell, and obtains the operating condition adaptive battery cell health index according to the real-time operating condition parameters of the battery cell and the battery cell health assessment model; obtains the predicted thermal runaway risk level according to the operating condition adaptive battery cell health index and the mapping relationship between the battery cell health index and the thermal runaway risk level; determines whether the predicted thermal runaway risk level exceeds the preset threshold, and if so, triggers the early warning mechanism.

Images