Battery thermal runaway dynamic monitoring and early warning system and method based on multiple parameters

Abstract

The present invention discloses a battery thermal runaway dynamic monitoring and early warning system and method based on multiple parameters, and relates to the technical field of battery monitoring and early warning. The system discloses a thermal runaway monitoring module, a thermal runaway primary early warning module, a thermal runaway internal risk analysis module, and a thermal runaway secondary early warning module. The thermal runaway internal risk analysis module and the thermal runaway secondary early warning module are provided. After starting the thermal runaway primary early warning, a battery simulation model is constructed, and the actual characteristics of the characterization of various types of thermal runaway characterization data are combined to simulate different internal defect conditions of the battery under the above characterization conditions, and an in-depth thermal runaway risk analysis is performed on each internal defect condition of the battery, and a comprehensive analysis is performed on the thermal runaway characterization condition and the thermal runaway internal condition of the battery.

Description

Technical Field

[0001] The present invention relates to the technical field of battery monitoring and early warning, and more specifically, to a battery thermal runaway dynamic monitoring and early warning system and method based on multiple parameters. Background Art

[0002] With the rapid development of the new energy industry, batteries, as core components for energy storage and conversion, have been widely used in electric vehicles, energy storage systems, portable electronic devices, etc. However, batteries may also cause thermal runaway during normal use, which will not only cause a sharp decline in battery performance, but may also cause serious safety accidents such as fire and explosion, posing a major threat to the safety of people and property.

[0003] At present, although there are some battery monitoring systems on the market, most of these systems focus on monitoring the conventional parameters of the battery, such as real-time collection of characteristic parameters such as voltage, current, and temperature. Whether the battery is in thermal runaway is determined by whether the corresponding characteristic parameters are abnormal, but the abnormality of characteristic parameters such as voltage, current, and temperature is not necessarily caused by battery thermal runaway. This leads to the fact that the early warning of battery thermal runaway may not be accurate, and there are often problems of false alarms and missed alarms. The accuracy of the early warning needs to be improved.

[0004] Secondly, the real cause of battery thermal runaway is often closely related to the defect parameters inside the battery. However, the real internal defect parameters of the battery are difficult to monitor and collect. Therefore, how to accurately monitor and warn of thermal runaway of the battery is an urgent problem that needs to be solved. Summary of the invention

[0005] In view of the deficiencies in the prior art, the object of the present invention is to provide a battery thermal runaway dynamic monitoring and early warning system and method based on multiple parameters.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A multi-parameter battery thermal runaway dynamic monitoring and early warning system, including a thermal runaway monitoring module, a thermal runaway primary early warning module, a thermal runaway internal risk analysis module, and a thermal runaway secondary early warning module;

[0008] The thermal runaway monitoring module collects various types of thermal runaway characterization data of the battery in real time during battery operation;

[0009] The thermal runaway primary warning module is used to periodically generate a thermal runaway prediction index Pnb of the battery, and determine whether to start the thermal runaway primary warning based on a comparison result between the thermal runaway prediction index and the thermal runaway prediction threshold index;

[0010] When the thermal runaway primary warning is started, the thermal runaway internal risk analysis module constructs a battery simulation model, obtains the actual characteristics of each type of thermal runaway characterization data in the current cycle, inputs the actual characteristics of each type of thermal runaway characterization data into the battery simulation model, and the battery simulation model completes S thermal runaway simulation analysis, thereby obtaining the thermal runaway internal risk value Dds of the battery;

[0011] The thermal runaway secondary warning module obtains the battery's thermal runaway comprehensive value based on the battery's thermal runaway prediction index Pnb and the thermal runaway internal risk value Dds, and determines whether to start the thermal runaway secondary warning based on a comparison result between the battery's thermal runaway comprehensive value and the thermal runaway comprehensive threshold. When it is determined that the thermal runaway secondary warning is to be started, the battery's thermal runaway secondary warning level is simultaneously determined.

[0012] Furthermore, a thermal runaway prediction index of the battery is generated periodically, specifically: after each cycle, characterization prediction features of various types of thermal runaway characterization data of the battery are obtained, the characterization prediction features are combined into a characterization prediction feature set, a thermal runaway analysis model is obtained, the characterization prediction feature set is used as input data of the thermal runaway analysis model, and the thermal runaway prediction index of the battery is output.

[0013] Furthermore, the characterization prediction features of a type of thermal runaway characterization data of a battery are obtained by the following steps: obtaining the actual characterization features of a type of thermal runaway characterization data for i consecutive cycles before the current time of the system, combining the actual characterization features of i cycles into a characterization cycle actual feature set, obtaining a characterization prediction model for the type of thermal runaway characterization data, using the characterization cycle actual feature set as input data of the characterization prediction model, and the characterization prediction model outputs the characterization prediction features of the type of thermal runaway characterization data of the battery.

[0014] Furthermore, the actual characteristics of a type of thermal runaway characterization data in a cycle are obtained by the following steps: collecting a type of thermal runaway characterization data of a battery in a cycle, performing feature extraction on the thermal runaway characterization data, and obtaining the actual characteristics of the type of thermal runaway characterization data in the cycle.

[0015] Furthermore, the battery simulation model construction process is as follows: select COMSOL battery simulation software, collect basic battery data, create a battery model in the battery simulation software based on the basic battery data, and assign corresponding parameters to the battery model, thereby constructing a battery simulation model.

[0016] Furthermore, the method for obtaining the internal thermal runaway risk value Dds of the battery is as follows: each time the battery simulation model completes a thermal runaway simulation analysis, the internal thermal runaway index of the thermal runaway simulation analysis is obtained, the internal thermal runaway indexes of S thermal runaway simulation analyses are matched pairwise into a simulation comparison group, and the stable risk value of each simulation comparison group is obtained. The stable risk values ​​of all simulation comparison groups are summed and averaged to obtain the internal thermal runaway risk value Dds of the battery.

[0017] Furthermore, each time the battery simulation model completes a thermal runaway simulation analysis, the thermal runaway internal index of the thermal runaway simulation analysis is obtained. Specifically, when the battery simulation model completes a thermal runaway simulation analysis, various internal defect parameters of the battery simulation model are obtained, feature extraction is performed on the various internal defect parameters to obtain defect features of the various internal defect parameters, feature analysis models corresponding to the various internal defect parameters are obtained, the defect features of the various internal defect parameters are respectively input into the corresponding feature analysis models, and the internal defect indexes Byt of the various internal defect parameters are output, y=1, 2, …, Y, Y is the total number of internal defect parameters, c is the number of the internal defect parameters, the internal defect coefficient is set to rt, t=1, 2, …, t, r1<r2<r3<…<rt, the internal defect threshold index is set, and when the internal defect index of the internal defect parameter is greater than the internal defect threshold index, the risk superposition times are increased once, and the risk superposition times are marked as Sng, and the formula is used The thermal runaway internal index Akh of this thermal runaway simulation analysis is obtained.

[0018] Furthermore, the stable risk value of the simulation comparison group is obtained as follows: the two thermal runaway internal indexes in the simulation comparison group are summed to obtain the thermal runaway risk sum index Lf, the two thermal runaway internal indexes in the simulation comparison group are differenced and the absolute value is taken to obtain the thermal runaway risk gap index Vm, and the formula is used The stable risk value of the simulation comparison group is obtained.

[0019] Furthermore, based on the battery's thermal runaway prediction index Pnb and the thermal runaway internal risk value Dds, the battery's thermal runaway comprehensive value is obtained, specifically: Using the formula The comprehensive thermal runaway value Kwu of the battery is obtained, where s1 is the thermal runaway prediction coefficient and s2 is the thermal runaway internal risk coefficient;

[0020] Determine the secondary thermal runaway warning level of the battery, specifically: perform difference calculation on the comprehensive thermal runaway value of the battery and the comprehensive thermal runaway threshold to obtain the secondary thermal runaway warning value Fj of the battery, set a total of j secondary thermal runaway warning levels, namely, secondary thermal runaway warning level 1, secondary thermal runaway warning level 2, ..., secondary thermal runaway warning level j-1, secondary thermal runaway warning level j, each range of secondary thermal runaway warning values ​​corresponds to a secondary thermal runaway warning level, the range of secondary thermal runaway warning values ​​includes (0, F1], (F1, F2], ..., (Fj-1, Fj], the thermal runaway risk of secondary thermal runaway warning level 1 is less than the thermal runaway risk of secondary thermal runaway warning level 2 is less than the thermal runaway risk of secondary thermal runaway warning level j-1 is less than the thermal runaway risk of secondary thermal runaway warning level j.

[0021] Furthermore, a multi-parameter battery thermal runaway dynamic monitoring and early warning method includes the following steps:

[0022] Step 1: During the operation of the battery, real-time collection of various types of thermal runaway characterization data of the battery;

[0023] Step 2: Periodically generate a thermal runaway prediction index Pnb of the battery, and determine whether to initiate a primary thermal runaway warning based on a comparison result between the thermal runaway prediction index and the thermal runaway prediction threshold index;

[0024] Step 3: When the primary warning of thermal runaway is started, a battery simulation model is constructed to obtain the actual characteristics of each type of thermal runaway characterization data in the current cycle, and the actual characteristics of each type of thermal runaway characterization data are input into the battery simulation model. The battery simulation model completes S thermal runaway simulation analysis, and then obtains the thermal runaway internal risk value Dds of the battery;

[0025] Step 4: Based on the battery's thermal runaway prediction index Pnb and thermal runaway internal risk value Dds, obtain the battery's thermal runaway comprehensive value. Based on the comparison result between the battery's thermal runaway comprehensive value and the thermal runaway comprehensive threshold, determine whether to start the thermal runaway secondary warning. When determining to start the thermal runaway secondary warning, simultaneously determine the battery's thermal runaway secondary warning level.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. The method of the present invention comprehensively analyzes the thermal runaway manifestation and internal conditions of the battery, and determines whether to start the secondary warning of thermal runaway and determines the secondary warning level of thermal runaway. By setting the primary warning of thermal runaway, the secondary warning of thermal runaway and the secondary warning level of thermal runaway, the warning standard of battery thermal runaway is dynamically set, which is conducive to the monitoring personnel to take corresponding measures for the battery in a targeted manner;

[0028] 2. Set up a thermal runaway monitoring module and a thermal runaway primary warning module to monitor and collect the thermal runaway characterization data of the battery in real time, and use the LSTM model to periodically predict the various thermal runaway characterization data of the battery. When predicting the possibility of thermal runaway of the battery, promptly start the corresponding thermal runaway primary warning, set up a thermal runaway internal risk analysis module and a thermal runaway secondary warning module. After starting the thermal runaway primary warning, build a battery simulation model, combine the actual characteristics of various types of thermal runaway characterization data, simulate the different internal defects of the battery under the above characterization conditions, and conduct in-depth thermal runaway risk analysis on each internal defect of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A method flow chart of a multi-parameter-based battery thermal runaway dynamic monitoring and early warning method;

[0030] Figure 2 This is a system module diagram of a multi-parameter-based battery thermal runaway dynamic monitoring and early warning system;

[0031] Figure 3 A flow chart for obtaining characterization prediction features for thermal runaway characterization data of a type of battery;

[0032] Figure 4 Flowchart for building a battery simulation model. DETAILED DESCRIPTION

[0033] Example 1: Reference Figure 1 , a multi-parameter battery thermal runaway dynamic monitoring and early warning method includes the following steps:

[0034] Step 1: During battery operation, collect various types of thermal runaway characterization data of the battery in real time.

[0035] Step 2: Periodically generate the thermal runaway prediction index Pnb of the battery, and determine whether to start the primary thermal runaway warning based on the comparison result between the thermal runaway prediction index and the thermal runaway prediction threshold index.

[0036] Step 3: When the primary warning of thermal runaway is started, a battery simulation model is constructed to obtain the actual characteristics of each type of thermal runaway characterization data in the current cycle, and the actual characteristics of each type of thermal runaway characterization data are input into the battery simulation model. The battery simulation model completes S thermal runaway simulation analyses, and then obtains the battery's thermal runaway internal risk value Dds.

[0037] Step 4: Based on the battery's thermal runaway prediction index Pnb and thermal runaway internal risk value Dds, obtain the battery's thermal runaway comprehensive value. Based on the comparison result between the battery's thermal runaway comprehensive value and the thermal runaway comprehensive threshold, determine whether to start the thermal runaway secondary warning. When determining to start the thermal runaway secondary warning, simultaneously determine the battery's thermal runaway secondary warning level.

[0038] The above method comprehensively analyzes the thermal runaway symptoms and internal conditions of the battery, and determines whether to start the thermal runaway secondary warning and determine the thermal runaway secondary warning level. By setting the thermal runaway primary warning, thermal runaway secondary warning and thermal runaway secondary warning level, the warning standard for battery thermal runaway is dynamically set, which is conducive to monitoring personnel taking corresponding measures for the battery in a targeted manner.

[0039] Example 2: Reference Figure 2-Figure 4 , a multi-parameter battery thermal runaway dynamic monitoring and early warning system, including a thermal runaway monitoring module, a thermal runaway primary early warning module, a thermal runaway internal risk analysis module, and a thermal runaway secondary early warning module.

[0040] Thermal runaway monitoring module: During battery operation, various types of thermal runaway characterization data of the battery are collected in real time (the data types of thermal runaway characterization data include battery temperature data, battery voltage data, battery current data, gas release data, battery deformation data, etc. Thermal runaway characterization data is data that is closely related to battery thermal runaway and can be collected by sensors and other equipment).

[0041] Thermal runaway primary warning module: periodically generates the thermal runaway prediction index Pnb of the battery, sets the thermal runaway prediction threshold index (the thermal runaway prediction threshold index is the preset index of the system), and starts the thermal runaway primary warning when the thermal runaway prediction index of the battery is greater than or equal to the thermal runaway prediction threshold index. When the thermal runaway prediction index of the battery is less than the thermal runaway prediction threshold index, no further processing is performed.

[0042] The thermal runaway prediction index of the battery is generated periodically. Specifically, after each cycle, the characterization prediction features of various types of thermal runaway characterization data of the battery are obtained, the characterization prediction features are combined into a characterization prediction feature set, the thermal runaway analysis model is obtained, the characterization prediction feature set is used as the input data of the thermal runaway analysis model, and the thermal runaway prediction index of the battery is output.

[0043] The characterization prediction features of a type of thermal runaway characterization data of a battery are obtained by the following steps: obtaining the actual characterization features of a type of thermal runaway characterization data for i consecutive cycles before the current time of the system (taking the battery temperature data as an example, the actual characterization features of the battery temperature data for i consecutive cycles are collected), combining the actual characterization features of i cycles into a characterization cycle actual feature set, obtaining a characterization prediction model for this type of thermal runaway characterization data, using the characterization cycle actual feature set as input data for the characterization prediction model, and the characterization prediction model outputs the characterization prediction features of this type of thermal runaway characterization data of the battery.

[0044] The actual characteristics of a type of thermal runaway characterization data in a cycle are obtained through the following steps: collect a type of thermal runaway characterization data of the battery in a cycle (taking battery temperature data as an example, collect all battery temperature data in a cycle), perform feature extraction on the thermal runaway characterization data (feature extraction includes statistical feature extraction, frequency domain feature extraction and other steps), and obtain the actual characteristics of the type of thermal runaway characterization data in the cycle.

[0045] Each type of thermal runaway characterization data corresponds to a characterization prediction model. All characterization prediction models are constructed based on the LSTM model. The difference between different characterization prediction models lies only in the difference in training data. In this embodiment, taking battery temperature data as an example, the construction process of the characterization prediction model of battery temperature data is disclosed: the actual feature sets of the characterization cycles of multiple battery temperature data are collected, and an LSTM model is constructed. The actual feature sets of the characterization cycles of the battery temperature data are used as training data for the LSTM model. A characterization prediction feature is assigned to each training data. The characterization prediction feature is the characterization feature for predicting the battery temperature data of the next cycle. The training data is divided into a training set, a validation set, and a test set according to a set ratio of 4:1:1. The training set, validation set, and test set are trained. After the training is completed, a characterization prediction model for the battery temperature data is constructed.

[0046] The construction process of the characterization prediction model for battery voltage data is as follows: if it is a characterization prediction model for battery voltage data, then the actual feature sets of the characterization cycles of multiple battery voltage data are collected, and the characterization prediction features are the characterization features for predicting the battery voltage data of the next cycle. The rest of the construction process is consistent with the construction process of the characterization prediction model for battery temperature data.

[0047] The thermal runaway analysis model is constructed based on the neural network model. The construction process of the thermal runaway analysis model is as follows: collect multiple characterization prediction feature sets, construct a neural network model, use the characterization prediction feature sets as training data for the neural network model, assign a thermal runaway prediction index to each training data, and the value range of the thermal runaway prediction index is (0.1~4.0). The larger the value of the thermal runaway prediction index, the greater the possibility of thermal runaway in the next cycle of the battery, and the smaller the value of the thermal runaway prediction index, the smaller the possibility of thermal runaway in the next cycle of the battery. The training data is divided into a training set and a validation set according to a set ratio of 5:1, and the neural network is iteratively trained on the training set and the validation set. After the training is completed, the thermal runaway analysis model is constructed.

[0048] A thermal runaway monitoring module and a thermal runaway primary warning module are set up to monitor and collect the thermal runaway characterization data of the battery in real time, and the LSTM model is used to periodically predict the various thermal runaway characterization data of the battery. When the possibility of thermal runaway of the battery is predicted, the corresponding thermal runaway primary warning is started in time.

[0049] Thermal runaway internal risk analysis module: When the primary warning of thermal runaway is started, a battery simulation model is constructed to obtain the actual characteristics of each type of thermal runaway characterization data in the current cycle, and the actual characteristics of each type of thermal runaway characterization data are input into the battery simulation model. The battery simulation model completes S thermal runaway simulation analysis (the actual characteristics of each type of thermal runaway characterization data are input into the battery simulation model. The battery simulation model will add internal defect parameters to the battery simulation model according to the actual characteristics of each type of thermal runaway characterization data. The internal defect parameters include internal short circuit parameters (including short circuit position, short circuit resistance, etc.), plate aging parameters (including plate deformation degree, plate internal resistance, plate active material desorption, etc.), and plate aging parameters (including plate deformation degree, plate internal resistance, plate active material desorption, etc.). After the internal defect parameters are adjusted, the battery simulation model simulates the actual work. After the actual work is simulated, when the various types of thermal runaway characterization data of the battery simulation model are consistent with the various types of thermal runaway characterization data of the battery, it means that a thermal runaway simulation analysis is completed. The internal defect parameters added in each completed thermal runaway simulation analysis are not completely consistent, so as to ensure that the internal defect conditions encountered in each thermal runaway simulation analysis are inconsistent, but the various types of thermal runaway characterization data generated are consistent. In summary, when the battery has different internal defects, the various types of thermal runaway characterization data may be consistent).

[0050] Each time the battery simulation model completes a thermal runaway simulation analysis, the thermal runaway internal index of the thermal runaway simulation analysis is obtained, the thermal runaway internal indexes of S thermal runaway simulation analyses are matched pairwise into a simulation comparison group, and the stable risk value of each simulation comparison group is obtained. The stable risk values ​​of all simulation comparison groups are summed and averaged to obtain the thermal runaway internal risk value Dds of the battery.

[0051] The stable risk value of the simulation comparison group is obtained as follows: the two thermal runaway internal indexes in the simulation comparison group are summed to obtain the thermal runaway risk sum index Lf, the two thermal runaway internal indexes in the simulation comparison group are differenced and the absolute value is taken to obtain the thermal runaway risk gap index Vm, and the formula is used The stable risk value of the simulation comparison group is obtained.

[0052] The process of building a battery simulation model is as follows: select battery simulation software such as COMSOL, collect basic battery data, create a battery model in the battery simulation software based on the basic battery data, and assign corresponding parameters to the battery model. The parameters include design parameters (such as battery size, electrode properties, etc.), physical parameters (such as electrode material properties, electrolyte conductivity, etc.), electrochemical parameters (such as electrochemical reaction parameters of reactive ions at the electrode interface, etc.), etc., and then construct a battery simulation model. The battery simulation model can simulate the actual operation of the battery.

[0053] Each time the battery simulation model completes a thermal runaway simulation analysis, the thermal runaway internal index of the thermal runaway simulation analysis is obtained. Specifically, when the battery simulation model completes a thermal runaway simulation analysis, various internal defect parameters of the battery simulation model are obtained (the internal defect parameters include internal short circuit parameters, plate aging parameters, battery water loss parameters, diaphragm defect parameters, etc.), and features are extracted for various internal defect parameters to obtain defect features of various internal defect parameters, and feature analysis models corresponding to various internal defect parameters are obtained. The defect features of various internal defect parameters are respectively input into the corresponding feature analysis models, and the internal defect indexes Byt, y=1, 2, ..., Y, Y is the total number of internal defect parameters, and c is the internal defect. The number of the parameter, set the internal defect coefficient to rt, t=1, 2, …, t, r1<r2<r3<…<rt, each internal defect coefficient corresponds to an internal defect index within a range, the value range of the internal defect index includes (0, By1], (By1, By2], …, (Byt-1, Byt], when Byt∈(0, By1], the internal defect coefficient is r1, set the internal defect threshold index (the internal defect threshold index is the preset index of the system), when the internal defect index of the internal defect parameter is greater than the internal defect threshold index, increase the risk superposition times by one, when the internal defect index of the internal defect parameter is less than or equal to the internal defect threshold index, no further processing is performed, and the risk superposition times are marked as Sng, using the formula The thermal runaway internal index Akh of this thermal runaway simulation analysis is obtained.

[0054] Each internal defect parameter corresponds to a feature analysis model, and all feature analysis models are constructed based on a neural network model. The difference between different feature analysis models lies only in the difference in training data. In this embodiment, taking the internal short-circuit parameter as an example, the construction process of the feature analysis model of the internal short-circuit parameter is disclosed: the defect features of multiple internal short-circuit parameters are collected, and a neural network model is constructed. The defect features of the internal short-circuit parameters are used as training data for the neural network model, and an internal defect index is assigned to each training data. The value range of the internal defect index is (1.1~5.0). The larger the value of the internal defect index, the higher the risk of battery thermal runaway caused by the internal short-circuit parameter. The smaller the value of the internal defect index, the lower the risk of battery thermal runaway caused by the internal short-circuit parameter. The training data is divided into a training set and a validation set according to a set ratio of 4:1, and the neural network is iteratively trained on the training set and the validation set. After the training is completed, a feature analysis model of the internal short-circuit parameter is constructed.

[0055] The construction process of the characteristic analysis model of the plate aging parameters is as follows: the defect characteristics of multiple plate aging parameters are collected. The larger the value of the internal defect index, the higher the risk of battery thermal runaway caused by the plate aging parameters; the smaller the value of the internal defect index, the lower the risk of battery thermal runaway caused by the plate aging parameters. The rest of the construction process is consistent with the construction process of the characteristic analysis model of the internal short circuit parameters.

[0056] Thermal runaway secondary warning module: Based on the battery's thermal runaway prediction index Pnb and thermal runaway internal risk value Dds, the battery's thermal runaway comprehensive value is obtained, and the thermal runaway comprehensive threshold is set (the thermal runaway comprehensive threshold is the system's preset threshold). When the battery's thermal runaway comprehensive value is greater than the thermal runaway comprehensive threshold, the thermal runaway secondary warning is started, and the battery's thermal runaway secondary warning level is determined simultaneously. When the battery's thermal runaway comprehensive value is less than or equal to the thermal runaway comprehensive threshold, no further processing is performed.

[0057] Based on the battery's thermal runaway prediction index Pnb and thermal runaway internal risk value Dds, the battery's thermal runaway comprehensive value is obtained: The comprehensive thermal runaway value Kwu of the battery is obtained, where s1 is the thermal runaway prediction coefficient, s2 is the thermal runaway internal risk coefficient, the value of s1 is 0.71, and the value of s2 is 0.79.

[0058] Determine the secondary thermal runaway warning level of the battery, specifically: perform difference calculation on the comprehensive thermal runaway value of the battery and the comprehensive thermal runaway threshold to obtain the secondary thermal runaway warning value Fj of the battery, set a total of j secondary thermal runaway warning levels, namely, secondary thermal runaway warning level 1, secondary thermal runaway warning level 2, ..., secondary thermal runaway warning level j-1, secondary thermal runaway warning level j, each range of secondary thermal runaway warning values ​​corresponds to a secondary thermal runaway warning level, and the range of secondary thermal runaway warning values ​​includes (0, F1], (F1, F2], …, (Fj-1, Fj], if the thermal runaway secondary warning value ∈ (0, F1], it is determined to be thermal runaway secondary warning level 1, the thermal runaway risk of thermal runaway secondary warning level 1 is less than the thermal runaway risk of thermal runaway secondary warning level 2 is less than the thermal runaway risk of thermal runaway secondary warning level j-1 is less than the thermal runaway risk of thermal runaway secondary warning level j; the monitoring personnel can determine the measures to be taken for the battery according to the warning mode (thermal runaway primary warning or thermal runaway secondary warning) and the warning level.

[0059] A thermal runaway internal risk analysis module and a thermal runaway secondary warning module are set up. After the primary warning of thermal runaway is started, a battery simulation model is constructed, and the actual characteristics of the characterization of various types of thermal runaway characterization data are combined to simulate the different internal defects of the battery under the above characterization conditions, and an in-depth thermal runaway risk analysis is performed on each internal defect of the battery. The thermal runaway characterization and internal conditions of the thermal runaway of the battery are comprehensively analyzed, and it is determined whether to start the secondary warning of thermal runaway and determine the secondary warning level of thermal runaway. By setting the primary warning of thermal runaway, the secondary warning of thermal runaway and the secondary warning level of thermal runaway, the warning standard of battery thermal runaway is dynamically set, which is conducive to the monitoring personnel to take corresponding measures for the battery in a targeted manner.

[0060] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0061] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0062] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0063] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multi-parameter battery thermal runaway dynamic monitoring and early warning system, characterized in that: Including thermal runaway monitoring module, thermal runaway primary warning module, thermal runaway internal risk analysis module, thermal runaway secondary warning module; The thermal runaway monitoring module collects various types of thermal runaway characterization data of the battery in real time during battery operation; The thermal runaway primary warning module is used to periodically generate a thermal runaway prediction index Pnb of the battery, and determine whether to start the thermal runaway primary warning based on a comparison result between the thermal runaway prediction index and the thermal runaway prediction threshold index; When the thermal runaway primary warning is started, the thermal runaway internal risk analysis module constructs a battery simulation model, obtains the actual characteristics of each type of thermal runaway characterization data in the current cycle, inputs the actual characteristics of each type of thermal runaway characterization data into the battery simulation model, and the battery simulation model completes S thermal runaway simulation analysis, thereby obtaining the thermal runaway internal risk value Dds of the battery; The thermal runaway secondary warning module obtains the battery's thermal runaway comprehensive value based on the battery's thermal runaway prediction index Pnb and the thermal runaway internal risk value Dds, and determines whether to start the thermal runaway secondary warning based on a comparison result between the battery's thermal runaway comprehensive value and the thermal runaway comprehensive threshold. When it is determined that the thermal runaway secondary warning is to be started, the battery's thermal runaway secondary warning level is simultaneously determined.

2. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 1 is characterized in that: The thermal runaway prediction index of the battery is generated periodically. Specifically, after each cycle, the characterization prediction features of various types of thermal runaway characterization data of the battery are obtained, the characterization prediction features are combined into a characterization prediction feature set, the thermal runaway analysis model is obtained, the characterization prediction feature set is used as the input data of the thermal runaway analysis model, and the thermal runaway prediction index of the battery is output.

3. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 2 is characterized in that: The characterization prediction features of a type of thermal runaway characterization data of a battery are obtained by the following steps: obtaining the actual characterization features of a type of thermal runaway characterization data for i consecutive cycles before the current time of the system, combining the actual characterization features of i cycles into a characterization cycle actual feature set, obtaining a characterization prediction model for this type of thermal runaway characterization data, using the characterization cycle actual feature set as input data of the characterization prediction model, and the characterization prediction model outputs the characterization prediction features of this type of thermal runaway characterization data of the battery.

4. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 3 is characterized in that: The actual characteristics of a type of thermal runaway characterization data in a cycle are obtained by the following steps: collecting a type of thermal runaway characterization data of a battery in a cycle, performing feature extraction on the thermal runaway characterization data, and obtaining the actual characteristics of the type of thermal runaway characterization data in the cycle.

5. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 1 is characterized in that: The process of building a battery simulation model is as follows: select COMSOL battery simulation software, collect basic battery data, create a battery model in the battery simulation software based on the basic battery data, assign corresponding parameters to the battery model, and then build a battery simulation model.

6. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 1 is characterized in that: The method for obtaining the internal thermal runaway risk value Dds of the battery is as follows: each time the battery simulation model completes a thermal runaway simulation analysis, the internal thermal runaway index of the thermal runaway simulation analysis is obtained, the internal thermal runaway indexes of S thermal runaway simulation analyses are matched pairwise into a simulation comparison group, and the stable risk value of each simulation comparison group is obtained. The stable risk values ​​of all simulation comparison groups are summed and averaged to obtain the internal thermal runaway risk value Dds of the battery.

7. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 6 is characterized in that: Each time the battery simulation model completes a thermal runaway simulation analysis, the thermal runaway internal index of the thermal runaway simulation analysis is obtained. Specifically, when the battery simulation model completes a thermal runaway simulation analysis, various internal defect parameters of the battery simulation model are obtained, and feature extraction is performed on the various internal defect parameters to obtain defect features of the various internal defect parameters, and feature analysis models corresponding to the various internal defect parameters are obtained. The defect features of the various internal defect parameters are respectively input into the corresponding feature analysis models, and the internal defect indexes Byt of the various internal defect parameters are output, y=1, 2, …, Y, Y is the total number of internal defect parameters, y is the number of the internal defect parameters, the internal defect coefficient is set to rt, t=1, 2, …, t, r1<r2<r3<…<rt, the internal defect threshold index is set, and when the internal defect index of the internal defect parameter is greater than the internal defect threshold index, the risk superposition times are increased once, and the risk superposition times are marked as Sng, and the formula is used. The thermal runaway internal index Akh of this thermal runaway simulation analysis is obtained.

8. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 6 is characterized in that: The stable risk value of the simulation comparison group is obtained as follows: the two thermal runaway internal indexes in the simulation comparison group are summed to obtain the thermal runaway risk sum index Lf, the two thermal runaway internal indexes in the simulation comparison group are differenced and the absolute value is taken to obtain the thermal runaway risk gap index Vm, and the formula is used The stable risk value of the simulation comparison group is obtained.

9. The multi-parameter-based battery thermal runaway dynamic monitoring and early warning system according to claim 8, characterized in that: Based on the battery's thermal runaway prediction index Pnb and thermal runaway internal risk value Dds, the battery's thermal runaway comprehensive value is obtained: The comprehensive thermal runaway value Kwu of the battery is obtained, where s1 is the thermal runaway prediction coefficient and s2 is the thermal runaway internal risk coefficient; Determine the secondary thermal runaway warning level of the battery, specifically: perform difference calculation on the comprehensive thermal runaway value of the battery and the comprehensive thermal runaway threshold to obtain the secondary thermal runaway warning value Fj of the battery, set a total of j secondary thermal runaway warning levels, namely, secondary thermal runaway warning level 1, secondary thermal runaway warning level 2, ..., secondary thermal runaway warning level j-1, secondary thermal runaway warning level j, each range of secondary thermal runaway warning values ​​corresponds to a secondary thermal runaway warning level, the range of secondary thermal runaway warning values ​​includes (0, F1], (F1, F2], ..., (Fj-1, Fj], the thermal runaway risk of secondary thermal runaway warning level 1 is less than the thermal runaway risk of secondary thermal runaway warning level 2 is less than the thermal runaway risk of secondary thermal runaway warning level j-1 is less than the thermal runaway risk of secondary thermal runaway warning level j.

10. A multi-parameter battery thermal runaway dynamic monitoring and early warning method, applied to a multi-parameter battery thermal runaway dynamic monitoring and early warning system according to any one of claims 1 to 9, characterized in that: The steps include: Step 1: During the operation of the battery, real-time collection of various types of thermal runaway characterization data of the battery; Step 2: Periodically generate a thermal runaway prediction index Pnb of the battery, and determine whether to initiate a primary thermal runaway warning based on a comparison result between the thermal runaway prediction index and the thermal runaway prediction threshold index; Step 3: When the primary warning of thermal runaway is started, a battery simulation model is constructed to obtain the actual characteristics of each type of thermal runaway characterization data in the current cycle, and the actual characteristics of each type of thermal runaway characterization data are input into the battery simulation model. The battery simulation model completes S thermal runaway simulation analysis, and then obtains the thermal runaway internal risk value Dds of the battery; Step 4: Based on the battery's thermal runaway prediction index Pnb and thermal runaway internal risk value Dds, obtain the battery's thermal runaway comprehensive value. Based on the comparison result between the battery's thermal runaway comprehensive value and the thermal runaway comprehensive threshold, determine whether to start the thermal runaway secondary warning. When determining to start the thermal runaway secondary warning, simultaneously determine the battery's thermal runaway secondary warning level.

Images