Lithium ion battery thermal safety evaluation system and method

Abstract

The invention relates to the technical field of lithium batteries, in particular to a thermal safety evaluation system and method for a lithium ion battery, and the method comprises the following steps: 1, collecting multi-dimensional data in the operation process of the lithium battery, and processing the multi-dimensional data as input parameters; 2, judging abuse types according to the input parameters; and 3, completing battery thermal safety state evaluation by adopting different thermal safety evaluation models according to different abuse types. According to the method, the safety state of the battery can be evaluated in real time by associating the multi-dimensional characteristic parameters of the lithium battery, the model is utilized to classify the safety levels of the battery, and early warning information can be provided for the occurrence and evolution of the thermal runaway behavior of the battery.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium batteries, and in particular to a lithium ion battery thermal safety assessment system and method. Background Art

[0002] Lithium-ion batteries have been widely used in many fields, such as consumer electronics, automobiles, and grid energy storage. However, the thermal safety issues that accompany their development have become a major obstacle to their promotion and application. The abuse of lithium-ion batteries in mechanical, electrical, and thermal environments may lead to severe thermal runaway. This runaway usually originates from side reactions inside the battery, causing heat accumulation, which eventually triggers a series of irreversible heat-generating processes, which may lead to serious consequences such as fire or explosion.

[0003] At present, the evaluation technology for the thermal safety of lithium-ion batteries is still relatively scarce. Existing evaluation methods mainly focus on thermal safety testing of batteries or risk probability assessment for thermal abuse, such as the invention patent: A method for thermal safety analysis of lithium batteries (CN 112098861 A). This patent uses a differential scanning calorimeter to analyze the thermal safety characteristics of the internal structure of lithium batteries, and performs simple analysis on each component of the lithium battery such as the positive electrode, negative electrode, and diaphragm material, and directly divides the dangerous degradation threshold. In contrast, this patent lacks a scientific and quantitative evaluation method, and does not have the ability to update and optimize the method model in real time based on real-time sample data. Another example is the invention patent: A method for assessing the safety risk of thermal abuse of lithium battery packs based on multi-physics field simulation (CN 114692244 A). This patent generates operating conditions and battery sample data through mathematical methods, and then evaluates the risk probability of thermal abuse of lithium batteries by establishing a safety critical value. Compared with this patent, these methods are based on ideal operating condition data generated by battery factory test data or mathematical model simulation. However, due to factors such as battery aging, consistency, and various complex external working conditions, these test conditions may be quite different from actual usage scenarios, resulting in the inability of the evaluation results to accurately reflect the battery's immediate thermal safety, and the evaluation methods are mostly based on simple threshold judgments. In addition, the current risk probability assessment also lacks a multi-dimensional comprehensive analysis of the degree of harm of lithium batteries under different degrees of abuse, and cannot demonstrate the multi-dimensional comprehensive immediate thermal safety of lithium batteries.

[0004] With the continuous advancement of battery technology and the diversification of application scenarios, the requirements for battery thermal safety are also increasing. Future research needs to introduce advanced sensing technology and data analysis methods to better understand the chemical and physical changes inside the battery and provide more accurate safety assessments. Improving the thermal safety of lithium-ion batteries requires not only improving existing materials and designs, but also strengthening the dynamic monitoring and evaluation of batteries under different environments. This will help ensure the safety and reliability of lithium-ion batteries in various applications, thereby promoting their wider application and development. Summary of the invention

[0005] The present invention provides a lithium-ion battery thermal safety assessment system and method, which aims to solve the problem that the thermal safety assessment of lithium-ion batteries is unclear. Traditional methods are temporarily unable to solve the safety assessment of lithium batteries at different stages under different abuse conditions and cannot determine their degree of danger.

[0006] A lithium ion battery thermal safety assessment method according to the present invention comprises the following steps:

[0007] Step 1: Collect multi-dimensional data during the operation of the lithium battery and process it as input parameters;

[0008] Step 2: Identify the abuse type based on the input parameters;

[0009] Step 3: Use different thermal safety assessment models to complete the thermal safety status assessment according to different abuse types.

[0010] Preferably, in step 1, specifically:

[0011] 1.1) Deploy multi-parameter battery data acquisition sensors at key locations on the battery surface to collect data;

[0012] 1.2) Obtain the cycle aging data of the battery cell, including real-time voltage, current, temperature, pressure, capacity decay, and internal resistance;

[0013] 1.3) Process the data to obtain the temperature time change rate, pressure time change rate, capacity retention rate, and voltage time change rate data;

[0014] 1.4) Use Kalman filter algorithm to smooth the data and analyze it;

[0015] 1.5) Take the processed data as input parameters.

[0016] Preferably, in step 2, specifically:

[0017] 2.1) After receiving the input parameters, if T>60℃ or dF / dt>25kPa / s, it means that the battery is in an unsafe state, and the input parameters are combined with the experience learning model for abuse analysis. Otherwise, the battery is considered safe and no abuse analysis is performed; where T represents the battery temperature, F represents the battery expansion pressure, and t represents time;

[0018] 2.2) If dT / dt≥1℃ / s, the warning system is directly reported and the abuse judgment is exited; if 0.02℃ / s<dT / dt<1℃ / s, the abuse type is matched and judged in the thermal abuse judgment model, electrical abuse judgment model and mechanical abuse judgment model;

[0019] 2.3) After empirical logic judgment of the data based on the three discrimination models, the historical expert database is transferred to assist in the judgment. The data of the historical expert database comes from the changes in thermal safety data of the same type of batteries under different abuse conditions. Based on the historical experimental data, the judgment results are weighted respectively;

[0020] 2.4) Distinguish the logical judgment intervals of the three discriminant models, and select the abuse type corresponding to the corresponding machine learning training interval as the judgment result;

[0021] Here, the softmax function is used to perform logistic regression classification on the three types of abuse;

[0022]

[0023] Among them C k represents the category, k is 3, x represents the input feature vector, Z k Represents the linear combination of the feature vector x under the kth category.

[0024] Preferably, in step 2.3), specifically:

[0025] In the electrical abuse discrimination model, if H2 is detected first and the voltage V is monitored to reach ±15% of the battery's specified cut-off voltage, it is judged as electrical abuse; a 0-1 hazard value is assigned based on the voltage increase or attenuation;

[0026] In the thermal abuse discrimination model, if organic gas precipitation is detected first and the ambient temperature of the battery is above 100°C, it is judged as thermal abuse; according to the ratio of organic gas to CO concentration, the higher the organic gas concentration, the greater the danger, and a 0-1 hazard value is assigned;

[0027] In the mechanical abuse discrimination model, if it is first detected that the pressure sensor reading is greater than 1MPa due to external force applied to the battery, and the time for the battery temperature to rise from the normal operating temperature to above 60°C is less than 5s or the battery voltage drops to less than 50% of the cut-off discharge voltage, it is judged as mechanical abuse; according to the degree of change of the dT / dt and dF / dt values, a hazard value of 0-1 is assigned.

[0028] Preferably, in step 3, the thermal safety assessment model includes an electrical abuse model, a thermal abuse model and a mechanical abuse model;

[0029] First, the correlation between the model input parameters and the predicted values ​​is analyzed by the Pearson correlation coefficient, with a value range of [-1, 1]. The correlation strength references are 0.8-1.0, 0.6-0.8, 0.4-0.6, 0.2-0.4, and 0.0-0.2, which are highly correlated, strongly correlated, moderately correlated, weakly correlated, and extremely weakly correlated, respectively. The following formula is used for judgment:

[0030]

[0031] ρ represents the correlation coefficient, x1 and x2 represent two different samples, cov is the covariance, Dx is the variance, which indicates the degree of deviation between the variable and the expectation; Ex is the expectation, that is, the average value of the variable;

[0032] For the electrical abuse model, voltage is a significant feature. The model determination features include voltage, temperature, resistance, and gas, which evaluate the thermal safety status of lithium batteries.

[0033] First, each parameter is estimated as an independent estimator, and finally the prediction results of each estimator are regressed and integrated through the combiner; specifically, the final regression result F can be expressed as the average value of all estimator results, that is:

[0034]

[0035] f i is the prediction result of the i-th estimator, and n is the number of estimators;

[0036] To train each estimator, sampling with replacement is performed from the running or experimental data with a sample size of N, repeated N times to form N training subsets; these samples are used to train independent estimators and become the training data of the estimator nodes;

[0037] In a sample with M attributes, each estimator randomly selects m attributes when splitting its node, where m is much smaller than M. The node selects one of the m attributes as the best split attribute, and the splitting process continues until the node splitting no longer improves the model performance. In this way, a large number of estimators are constructed to form a complete electrical abuse assessment model.

[0038] For the thermal abuse model, the temperature change rate is a significant feature. The model determination features include temperature, voltage, and gas composition to evaluate the thermal safety status of lithium batteries.

[0039] After the estimator splits, due to the data error of the thermal abuse model, it is necessary to use the optimization algorithm to tune the hyperparameters of the model to find the optimal combination, as follows:

[0040] Here, K-fold cross-certification is used to divide the data set into K equal parts, one of which is used as a validation set, and the remaining K-1 parts are used as training sets. According to this logic, it is repeated K times to obtain K models and corresponding errors. The average of all errors is calculated to obtain the generalization error E of the model, as shown in the following formula:

[0041]

[0042] At the same time, a smoother The covariance function is used as the covariance function for calculating the Gaussian process:

[0043]

[0044] σ 2 represents variance, r represents the distance difference between signals;

[0045] For the mechanical abuse model, pressure is a significant feature. The model judgment features include pressure, voltage, and temperature, and the thermal safety status of the lithium battery is evaluated. The evaluation method is the same as above. Considering the special dangers of mechanical abuse, if the battery temperature rise dT / dt during the evaluation process is greater than 0.5℃ / s, the early warning system will be directly linked to report a level 3 alarm.

[0046] Preferably, in step 3, the evaluation method is:

[0047] The thermal safety status of lithium batteries is graded based on the results of three thermal safety assessment models;

[0048] The evaluation safety interval is [0-1], referring to the battery health status SOH. When SOH<0.8, the battery needs to be replaced. Similarly, the value σ is defined. When the evaluation safety value is in the [σ-1] interval, it is considered that the temperature rise of the battery under abuse is relatively safe and controllable. When the battery evaluation safety calculation value is on [0-σ], it is considered that the battery temperature rises quickly under different abuse conditions, the safety is poor, and its thermal safety is uncontrollable.

[0049] In addition, the safety status of lithium batteries is divided into four levels: safety, level 1 warning, level 2 warning and level 3 warning, and thresholds σ1, σ2 and σ3 are set according to different types of batteries;

[0050] By comparing the assessment value range, the battery safety is rated; when the assessment safety value is in the [0-σ1] range, it is safe; when the assessment safety value is in the [σ1-σ2] range, it is a level one alarm; when the assessment safety value is in the [σ2-σ3] range, it is a level two alarm; when the assessment safety value is in the [σ3-1] range, it is a level three alarm.

[0051] The present invention provides a lithium ion battery thermal safety assessment system, which adopts the above-mentioned lithium ion battery thermal safety assessment method.

[0052] The present invention establishes a thermal safety model through coupled analysis of battery aging data and thermal runaway data, completes the quantitative evaluation of thermal safety of lithium-ion batteries, and ensures the safe use of batteries. The present invention can evaluate the current safety of lithium batteries through various related parameters, and use the model to classify the safety level of batteries, which can provide early warning information for the occurrence and evolution of thermal runaway behavior of batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The figure is a flow chart of a method for thermal safety assessment of a lithium-ion battery in an embodiment. DETAILED DESCRIPTION

[0054] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0055] Example

[0056] like Figure 1 As shown, this embodiment proposes a lithium-ion battery thermal safety assessment method, which performs a lithium-ion battery thermal safety assessment on a commercial lithium titanate battery (rated capacity 9Ah), and includes the following steps:

[0057] Step 1: Collect multi-dimensional data during the operation of the lithium battery and process it as input parameters;

[0058] In step 1, specifically:

[0059] 1.1) Deploy multi-parameter battery data acquisition sensors at key locations on the battery surface to collect data;

[0060] 1.2) Obtain the cycle aging data of the battery cell, which is obtained through the battery single charge and discharge test; including the instantaneous voltage, current, temperature, sensor pressure value, capacity at different aging cycles, resistance retention rate, etc.;

[0061] 1.3) Process the data to obtain data such as temperature time change rate, pressure time change rate, capacity retention rate, voltage time change rate, etc.;

[0062] 1.4) The data is smoothed using the Kalman filter algorithm to reduce errors and then analyzed;

[0063] 1.5) Take the processed data as input parameters.

[0064] Step 2: Identify the abuse type based on the input parameters;

[0065] In step 2, specifically:

[0066] 2.1) After receiving the input parameters, the temperature around the battery is measured to be 118°C, which means the battery is in an unsafe state. The input parameters are combined with the experience learning model for abuse analysis;

[0067] 2.2) At this time, the dT / dt obtained by analysis is 0.15℃ / s, and the abuse type is matched and judged in the thermal abuse discrimination model, electrical abuse discrimination model and mechanical abuse discrimination model;

[0068] At this time, the sensor first detected the CO gas component, and the temperature around the battery was greater than 100°C. The conformity of the three major abuse judgment models was comprehensively matched. The results were as follows: thermal abuse was 0.88, electrical abuse was 0.10, and mechanical abuse was 0.02. It was initially judged that thermal abuse had occurred, and the weight value was set to 0.7;

[0069] 2.3) Expert data and historical expert database are transferred in for auxiliary judgment. The data of the historical expert database comes from the aging data changes of the same type of batteries under different abuse conditions. Based on the historical experimental data, the current model is weighted with a weight value of 0.3;

[0070] 2.4) Distinguish the logical judgment intervals of the three discriminant models, and select the abuse type corresponding to the corresponding machine learning training interval as the judgment result;

[0071] Here, the softmax function is used to perform logistic regression classification on the three types of abuse;

[0072]

[0073] Among them C k represents the category, k is 3, x represents the input feature vector, Z k Represents the linear combination of the feature vector x under the kth category.

[0074] Substituting the processed battery parameter information into the model, the probability output is 0.85 for thermal abuse, 0.10 for electrical abuse, and 0.05 for mechanical abuse. After comprehensive weighting, the thermal abuse, electrical abuse, and mechanical abuse are 0.871, 0.1, and 0.029, respectively, indicating that the system has suffered from thermal abuse.

[0075] In step 3, since it is determined that the system is subjected to thermal abuse, the thermal abuse assessment model is called;

[0076] First, the correlation between the model input parameters and the safety assessment value is calculated and analyzed by the Pearson correlation coefficient. The value range is [-1, 1]. The correlation strength references are 0.8-1.0, 0.6-0.8, 0.4-0.6, 0.2-0.4, and 0.0-0.2, which are highly correlated, strongly correlated, moderately correlated, weakly correlated, and extremely weakly correlated, respectively. The following formula is used for judgment:

[0077]

[0078] ρ represents the correlation coefficient, x1 and x2 represent two different samples, cov is the covariance, Dx is the variance, which indicates the degree of deviation between the variable and the expectation; Ex is the expectation, that is, the average value of the variable.

[0079] For the thermal abuse model of lithium titanate battery in this example, the temperature change rate is a significant feature. The model determination features include temperature, voltage, resistance, and gas composition, which evaluate the thermal safety status of the lithium battery.

[0080] The temperature change rate, temperature, voltage, resistance, and CO concentration are estimated as independent estimators, and the prediction results of each estimator are finally regressed and integrated through a combiner. Specifically, the final regression result F can be expressed as the average value of all estimator results, that is:

[0081]

[0082] f is the prediction result of the i-th estimator, n is the number of estimators, here n is 5;

[0083] First, we train each estimator by sampling with replacement from 100 sample data, repeating it 100 times to form 100 training subsets. These samples are used to train independent estimators and become the training data of the estimator nodes to improve the robustness of the model.

[0084] In a sample with M attributes, each estimator randomly selects m attributes when splitting a node, where m is much smaller than M. The node selects one of the m attributes as the best split attribute, and the splitting process continues until the node split no longer significantly improves the model performance, which is manifested as less than 1% improvement in optimization compared to the previous one. In this way, we can build a large number of estimators to form a complete abuse assessment model.

[0085] Then, K-fold cross validation is used to divide the data set into 4 equal parts, one of which is used as a validation set and the remaining 3 as training sets. Following this logic, it is repeated 4 times to obtain 4 models and corresponding errors. The average of all errors is calculated to obtain the generalization error E of the model, as shown in the following formula:

[0086]

[0087] At the same time, a smoother The covariance function is used as the covariance function for calculating the Gaussian process:

[0088]

[0089] σ 2represents variance, r represents the distance difference between signals;

[0090] Finally, we used the mean absolute percentage error (MAPE), root mean square percentage error (RMSPE), and coefficient of determination (R 2 ) to evaluate the performance of the model.

[0091] Similarly, electrical abuse and mechanical abuse analysis are performed on the remaining cases; the results are shown in Table 1, which are the evaluation results of the electrical abuse model, mechanical abuse model, and thermal abuse model.

[0092] Table 1 Evaluation results of electrical abuse model, mechanical abuse model and thermal abuse model

[0093] Model Name MAPE(%) RMSPE(%) <![CDATA[R 2 ]]> Electrical Abuse Model 3.336 3.880 0.712 Mechanical Abuse Model 2.998 3.364 0.776 Thermal Abuse Model 3.228 3.625 0.737

[0094] From the data in Table 1, we can see that different models R 2 All are above 0.712, MAPE is below 3.336 and RMSPE is below 3.880, showing good predictability and evaluation.

[0095] Preferably, in step 3, the evaluation method is:

[0096] The thermal safety of lithium batteries is graded based on the results of three thermal safety assessment models;

[0097] The evaluation safety interval is [0-1], referring to the battery health status SOH. When SOH<0.8, the battery needs to be replaced. Similarly, the value σ is defined. When the evaluation safety value is in the [σ-1] interval, it is considered that the battery temperature rise is relatively safe and controllable under extreme abuse. When the battery evaluation safety calculation value is on [0-σ], it is considered that the battery temperature rises quickly under different abuse conditions, the safety is poor, and its thermal safety is uncontrollable.

[0098] In addition, the safety status level of the lithium battery is divided into four levels: safety, level 1 warning, level 2 warning and level 3 warning. For the battery in this embodiment, the thresholds σ1=0.25, σ2=0.58 and σ3=0.72 are set;

[0099] According to the interval where the model output evaluation value is 0.324, the battery issues a level 1 warning.

[0100] This embodiment provides a lithium-ion battery thermal safety assessment system, which adopts the above-mentioned lithium-ion battery thermal safety assessment method.

[0101] This embodiment establishes a thermal safety model through coupled analysis of battery aging data and thermal runaway data, completes the quantitative evaluation of thermal safety of lithium-ion batteries, and ensures the safe use of batteries. This embodiment can evaluate the current safety of lithium batteries through various related parameters, and use the model to classify the safety level of batteries, which can pave the way for timely warning of battery safety in the future.

[0102] The present invention and its embodiments are described schematically above, and the description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. Therefore, if a person skilled in the art is inspired by it and designs a structural method and an embodiment similar to the technical solution without creativity without departing from the purpose of the invention, they shall all fall within the protection scope of the present invention.

Claims

1. A method for thermal safety assessment of lithium-ion batteries, characterized in that: The following steps are involved: Step 1: Collect multi-dimensional data during the operation of the lithium battery and process it as input parameters; Step 2: Identify the abuse type based on the input parameters; Step 3: Use different thermal safety assessment models to complete the thermal safety status assessment according to different abuse types.

2. The method for thermal safety assessment of a lithium-ion battery according to claim 1, characterized in that: In step 1, specifically: 1.1) Deploy multi-parameter battery data acquisition sensors at key locations on the battery surface to collect data; 1.2) Obtain the cycle aging data of the battery cell, including real-time voltage, current, temperature, pressure, capacity decay, and internal resistance; 1.3) Process the data to obtain the temperature time change rate, pressure time change rate, capacity retention rate, and voltage time change rate data; 1.4) Use Kalman filter algorithm to smooth the data and analyze it; 1.5) Take the processed data as input parameters.

3. The method for thermal safety assessment of a lithium-ion battery according to claim 2, characterized in that: In step 2, specifically: 2.1) After receiving the input parameters, if T>60℃ or dF / dt>25kPa / s, it means that the battery is in an unsafe state, and the input parameters are combined with the experience learning model for abuse analysis. Otherwise, the battery is considered safe and no abuse analysis is performed; where T represents the battery temperature, F represents the battery expansion pressure, and t represents time; 2.2) If dT / dt≥1℃ / s, the warning system is directly reported and the abuse judgment is exited; if 0.02℃ / s<dT / dt<1℃ / s, the abuse type is matched and judged in the thermal abuse judgment model, electrical abuse judgment model and mechanical abuse judgment model; 2.3) After empirical logic judgment of the data based on the three discrimination models, the historical expert database is transferred to assist in the judgment. The data of the historical expert database comes from the changes in thermal safety data of the same type of batteries under different abuse conditions. Based on the historical experimental data, the judgment results are weighted respectively; 2.4) Distinguish the logical judgment intervals of the three discriminant models, and select the abuse type corresponding to the corresponding machine learning training interval as the judgment result; Here, the softmax function is used to perform logistic regression classification on the three types of abuse; Among them C k represents the category, k is 3, x represents the input feature vector, Z k Represents the linear combination of the feature vector x under the kth category.

4. The method for thermal safety assessment of a lithium-ion battery according to claim 3, characterized in that: In step 2.3), specifically: In the electrical abuse discrimination model, if H2 is detected first and the voltage V is monitored to reach ±15% of the battery's specified cut-off voltage, it is judged as electrical abuse; a 0-1 hazard value is assigned based on the voltage increase or attenuation; In the thermal abuse discrimination model, if organic gas precipitation is detected first and the ambient temperature of the battery is above 100°C, it is judged as thermal abuse; according to the ratio of organic gas to CO concentration, the higher the organic gas concentration, the greater the danger, and a 0-1 hazard value is assigned; In the mechanical abuse discrimination model, if it is first detected that the pressure sensor reading is greater than 1MPa due to external force applied to the battery, and the time for the battery temperature to rise from the normal operating temperature to above 60°C is less than 5s or the battery voltage drops to less than 50% of the cut-off discharge voltage, it is judged as mechanical abuse; according to the degree of change of the dT / dt and dF / dt values, a hazard value of 0-1 is assigned.

5. The method for thermal safety assessment of a lithium-ion battery according to claim 4, characterized in that: In step 3, the thermal safety assessment model includes an electrical abuse model, a thermal abuse model, and a mechanical abuse model; First, the correlation between the model input parameters and the predicted values ​​is analyzed by the Pearson correlation coefficient, with a value range of [-1, 1]. The correlation strength references are 0.8-1.0, 0.6-0.8, 0.4-0.6, 0.2-0.4, and 0.0-0.2, which are highly correlated, strongly correlated, moderately correlated, weakly correlated, and extremely weakly correlated, respectively. The following formula is used for judgment: ρ represents the correlation coefficient, x1 and x2 represent two different samples, cov is the covariance, Dx is the variance, which indicates the degree of deviation between the variable and the expectation; Ex is the expectation, that is, the average value of the variable; For the electrical abuse model, voltage is a significant feature. The model determination features include voltage, temperature, resistance, and gas, which evaluate the thermal safety status of lithium batteries. First, each parameter is estimated as an independent estimator, and finally the prediction results of each estimator are regressed and integrated through the combiner; specifically, the final regression result F can be expressed as the average value of all estimator results, that is: f i is the prediction result of the i-th estimator, and n is the number of estimators; To train each estimator, sampling with replacement is performed from the running or experimental data with a sample size of N, repeated N times to form N training subsets; These samples are used to train independent estimators and become the training data of the estimator nodes; In a sample with M attributes, each estimator node splits and randomly selects m attributes from them, where m is much smaller than M; The node selects one of the m attributes as the best split attribute, and the splitting process continues until the node splitting no longer improves the model performance. In this way, a large number of estimators are constructed to form a complete electrical abuse assessment model. For the thermal abuse model, the temperature change rate is a significant feature. The model determination features include temperature, voltage, and gas composition to evaluate the thermal safety status of lithium batteries. After the estimator splits, due to the data error of the thermal abuse model, it is necessary to use the optimization algorithm to tune the hyperparameters of the model to find the optimal combination, as follows: Here, K-fold cross-certification is used to divide the data set into K equal parts, one of which is used as a validation set, and the remaining K-1 parts are used as training sets. According to this logic, it is repeated K times to obtain K models and corresponding errors. The average of all errors is calculated to obtain the generalization error E of the model, as shown in the following formula: At the same time, a smoother The covariance function is used as the covariance function for calculating the Gaussian process: σ 2 represents variance, r represents the distance difference between signals; For the mechanical abuse model, pressure is a significant feature. The model judgment features include pressure, voltage, and temperature, and the thermal safety status of the lithium battery is evaluated. The evaluation method is the same as above. Considering the special dangers of mechanical abuse, if the battery temperature rise dT / dt during the evaluation process is greater than 0.5℃ / s, the early warning system will be directly linked to report a level 3 alarm.

6. The method for thermal safety assessment of a lithium-ion battery according to claim 5, characterized in that: In step 3, the evaluation method is: The thermal safety status of lithium batteries is graded based on the results of three thermal safety assessment models; The evaluation safety interval is [0-1], referring to the battery health status SOH. When SOH<0.8, the battery needs to be replaced. Similarly, the value σ is defined. When the evaluation safety value is in the [σ-1] interval, it is considered that the temperature rise of the battery under abuse is relatively safe and controllable. When the battery evaluation safety calculation value is on [0-σ], it is considered that the battery temperature rises quickly under different abuse conditions, the safety is poor, and its thermal safety is uncontrollable. In addition, the safety status of lithium batteries is divided into four levels: safety, level 1 warning, level 2 warning and level 3 warning, and thresholds σ1, σ2 and σ3 are set according to different types of batteries; By comparing the assessment value range, the battery safety is rated; when the assessment safety value is in the [0-σ1] range, it is safe; when the assessment safety value is in the [σ1-σ2] range, it is a level one alarm; when the assessment safety value is in the [σ2-σ3] range, it is a level two alarm; when the assessment safety value is in the [σ3-1] range, it is a level three alarm.

7. Lithium-ion battery thermal safety assessment system, characterized by: It adopts the lithium-ion battery thermal safety assessment method as described in any one of claims 1-6.

Images