Lithium battery thermal runaway risk early warning system based on fast charging scene

Abstract

The invention relates to the technical field of battery monitoring, in particular to a lithium battery thermal runaway risk early warning system based on a fast charging scene, which comprises a thermal diffusion monitoring module, a voltage response identification module, a voltage difference trend extraction module, an energy deviation judgment module and a risk level output module. According to the invention, through combining the direction matching relation of the temperature rise, the voltage and the voltage difference signal, the identification capability of the nonlinear cross trend in the fast charging process is optimized, and through combining the coupling analysis of the heat flux offset and the energy conversion proportion, the response sensitivity of the thermal runaway risk precursor signal is enhanced; risk levels are subjected to interval division through staged fluctuation intensity distribution characteristics, the risk assessment accuracy is improved, continuous monitoring of risk states in a time sequence is achieved in combination with trend continuity and multi-parameter matching intensity information, the operation safety and assessment accuracy under the high-rate charging condition are enhanced, and the risk assessment accuracy is improved. And the abnormal state judgment efficiency under the complex thermoelectric imbalance condition is improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of battery monitoring, in particular to a lithium battery thermal runaway risk early warning system based on a fast charging scene. BACKGROUND

[0002] The technical field of battery monitoring includes real-time detection and management of the operating state of various electrochemical batteries, mainly covering the collection and analysis of key parameters such as the voltage, current, temperature, state of charge and health status of the battery, focusing on quantitatively monitoring various indicators of the battery during operation through the establishment of a multi-parameter sensing measurement system, and dynamically evaluating the performance and safety of the battery by means of data acquisition devices, signal conversion circuits and state judgment logic, through the integration of electric signal detection, electric quantity estimation methods, environmental parameter processing methods and alarm judgment logic, etc., it is applied to scenes such as electric vehicles, energy storage systems and portable electronic devices, to ensure the reliability and safety of the battery during use. The lithium battery thermal runaway risk early warning system based on the fast charging scene refers to the situation that the lithium battery is in a thermal runaway risk due to rapid increase of internal temperature rise during high-rate charging. The real-time response of the cell monomer during charging is monitored by setting a voltage acquisition circuit, a current detection channel and a temperature sensing unit. The temperature rise rate is estimated by combining a heat change calculation model, and the abnormal trend is identified by comparing the threshold value. The potential thermal runaway characteristics are judged. A temperature distribution map is constructed by arranging multiple temperature sampling points on the surface and around the cell. A measurement circuit for voltage deviation judgment and an analysis channel for identifying current mutation are provided. Multiple electric signals are analyzed to construct risk identification basis. Relying on the signal acquisition path and micro processing chip of the battery pack, the risk is identified by using the temperature rise rate and voltage fluctuation in a fixed time window.

[0003] The traditional lithium battery thermal runaway risk early warning technology relies on the temperature and voltage comparison logic of the fixed threshold value in the process of battery fast charging monitoring. The parameters are independent of each other and lack dynamic coupling judgment, which is difficult to reflect the time delay and response asynchronization between multiple parameters. In the initial stage of thermal runaway, only the obvious temperature rise result can be monitored, and the energy accumulation trend cannot be identified. When the cell is in a short-term unbalanced charging or local polarization state, the temperature and voltage fluctuation may show reverse changes in different time periods, resulting in misjudgment of the risk level, alarm lag or false triggering problems, and affecting the reliability of the thermal runaway risk prediction. SUMMARY

[0004] The purpose of the present application is to solve the shortcomings in the prior art, and the lithium battery thermal runaway risk early warning system based on the fast charging scene is proposed.

[0005] In order to achieve the above object, the present application adopts the following technical scheme, the lithium battery thermal runaway risk early warning system based on fast charging scene comprises: The heat diffusion monitoring module calls the cell surface heat flux data, calculates the ratio difference of the heat flux change of adjacent regions, judges the change direction in the continuous period, identifies the position with the largest change amplitude in combination with the trend of the current and the last period ratio difference, and generates heat flux offset degree information; The voltage response identification module extracts the corresponding region temperature rise and voltage change data based on the heat flux offset degree information, calculates the directionality combination of the two, screens the reverse offset relationship, analyzes the synchronization degree of the combination sequence in combination with the time interval, and generates the thermal-electric response lag amplitude; The differential pressure trend extraction module calls the thermal-electric response lag amplitude, extracts the positive and negative electrode differential pressure data in the corresponding period, analyzes the change direction and growth amplitude difference, screens the same direction growth sequence in combination with the stability of growth and direction, extracts the differential pressure amplitude maximum point, and generates the differential pressure trend amplitude information; The energy offset determination module calls the differential pressure trend amplitude information, extracts the current and temperature rise change rate data, judges the synchronization of the ratio direction and the current change, analyzes the thermal energy response matching degree in combination with the current mutation amplitude and the temperature rise change rate, and generates the energy conversion offset information; The risk level output module calls the energy conversion offset information, extracts the change direction of multiple parameters in the same period, calculates the matching strength of the reverse change of temperature rise and voltage, calculates the change strength between the target trend and the energy conversion offset ratio in combination with the differential pressure direction consistency, and outputs the risk level division interval information according to the stage fluctuation distribution characteristics of the strength in the sequence.

[0006] As a further scheme of the present application, the heat flux offset degree information includes the heat diffusion main path range, the heat flux ratio difference change trend, and the heat flux abnormal concentration position, the thermal-electric response lag amplitude specifically refers to the voltage temperature rise directionality combination relationship, the directionality offset time sequence, and the change synchronization degree characteristics, the differential pressure trend amplitude information specifically refers to the differential pressure growth trend direction, the differential pressure maximum growth point, and the temperature rise coupling trend direction, the energy conversion offset information includes the current and temperature rise ratio change direction, the thermal-electric response synchronization trend, and the current mutation section, and the risk level division interval information specifically refers to the reverse change matching strength, the cross trend distribution stage, and the thermal-electric offset joint strength.

[0007] As a further scheme of the present application, the heat diffusion monitoring module comprises: The heat flux acquisition submodule calls the cell surface heat flux data, analyzes the heat flux change range in multiple direction regions of the cell surface, calculates the ratio difference of the heat flux change between adjacent regions, and generates the heat flux change interval; The ratio trend screening submodule calculates the ratio difference of the heat flux change of adjacent areas in each period based on the heat flux change interval, judges the change direction of the ratio difference in consecutive periods, combines the ratio difference change trend of the current period and the previous period, screens the area sequence with consistent trend, and obtains a ratio trend sequence; The main path determination submodule calls the ratio trend sequence, compares the ratio change amplitude of each area in the sequence, extracts the position with the largest amplitude as the heat diffusion main path range, and obtains the heat flux offset degree information.

[0008] As a further scheme of the present application, the voltage response recognition module comprises: The rate extraction analysis submodule extracts the cell voltage data and temperature rise change data of the corresponding area based on the heat flux offset degree information, analyzes the voltage increment value and temperature increment value in each period, obtains the voltage change rate and temperature rise rate data, combines the normalization processing, and obtains the rate sequence processing result; The direction matching judgment submodule calls the rate sequence processing result, analyzes the directional features of the temperature rise rate and the voltage change rate in each period, calculates the combined response relationship of the two in each period, screens the combined relationship of the reverse offset, obtains the time period and the corresponding index section formed by the reverse combination in consecutive periods, extracts the combined sequence formed by the continuous offset behavior, and obtains the reverse offset section sequence; The response lag calculation submodule calls the reverse offset section sequence, obtains the voltage change rate and temperature rise rate value of each period point in the sequence, calculates the degree of asynchronization in the time and amplitude dimensions, judges the overall lag response strength of the period section, and obtains the heat-electricity response lag amplitude.

[0009] As a further scheme of the present application, the pressure difference trend extraction module comprises: The pressure difference change analysis submodule calls the heat-electricity response lag amplitude, extracts the pressure difference data of the positive electrode and the negative electrode of the corresponding period of the cell, analyzes the change direction of the pressure difference in consecutive periods, and generates a pressure difference change direction sequence; The growth continuation screening submodule calculates the difference of the pressure difference growth amplitude of adjacent periods based on the pressure difference change direction sequence, judges the consistency of the change direction and locates the trend continuation section, screens the period sequence with continuous same-direction growth, and obtains same-direction growth section data; The trend amplitude determination submodule compares the pressure difference change amplitudes of all periods in the section according to the same-direction growth section data, extracts the period point with the largest pressure difference change amplitude, combines the temperature rise change direction data corresponding to the period, and obtains the pressure difference trend amplitude information.

[0010] As a further scheme of the present application, the energy offset determination module comprises: The ratio direction calculation sub-module extracts the current change rate and temperature rise change rate data in the corresponding period based on the pressure difference trend amplitude information, analyzes the ratio change direction of the current change rate and temperature rise change rate in the continuous period, and establishes a ratio direction sequence; The synchronous trend screening sub-module calls the ratio direction sequence, extracts the current change direction information in the corresponding period, calculates the change amplitude of the ratio of adjacent periods, judges the synchronism of the ratio change direction and the current change direction, screens the period sequence with synchronous rising in the continuous period, generates a synchronous rising trend section sequence in combination with the temperature rise change rate data in the section, and outputs the risk level division interval information according to the stage fluctuation distribution characteristics of the strength in the sequence. The response matching recognition sub-module extracts the current mutation amplitude and temperature rise change rate in the target section according to the synchronous rising trend section sequence, analyzes the adjustment offset strength between the current and the temperature rise change rate in the target period in combination with the pressure difference change rate, calculates the electrothermal response matching offset value, and obtains the energy conversion offset information in combination with the centralized distribution characteristics of the offset strength in each period.

[0011] As a further scheme of the present application, the risk level output module comprises: The parameter combination extraction sub-module calls the energy conversion offset information, extracts the temperature rise change rate, voltage change rate and pressure difference change direction data in the same period, analyzes the change direction of each parameter in the period sequence, and generates a parameter direction combination sequence; The reverse matching judgment sub-module calculates the matching strength of the reverse change of the temperature rise and the voltage based on the parameter direction combination sequence, judges the distribution characteristics of the multi-parameter cross trend in the time sequence in combination with the time consistency of the matching strength and the pressure difference direction continuation characteristics, and obtains multi-parameter trend distribution information; The level interval output sub-module calculates the change strength between the target trend and the energy conversion offset ratio according to the multi-parameter trend distribution information, and outputs the risk level division interval information according to the stage fluctuation distribution characteristics of the strength in the sequence.

[0012] As a further scheme of the present application, the process of outputting the risk level division interval information according to the stage fluctuation distribution characteristics of the strength in the sequence is specifically: forming a change strength sequence in chronological order according to the change strength, performing stage division on the change strength sequence based on adjacent extreme inflection points, and obtaining a stage set; For each stage, the stage fluctuation range, the stage average strength and the stage continuous rising proportion are calculated. The fluctuation threshold is set based on the median of the stage fluctuation range of all stages in the same change strength sequence, the strength threshold is set based on the median of the stage average strength of all stages, and the continuity threshold is set based on the median of the stage continuous rising proportion of all stages. The threshold matching is performed on the phase set according to the fluctuation threshold, the intensity threshold and the continuity threshold, and risk level division interval information meeting the threshold combination relationship is output.

[0013] Compared with the prior art, the application has the advantages and positive effects that: In the application, by combining the direction matching relationship of the temperature rise, the voltage and the pressure difference signal, the identification ability of the nonlinear cross trend in the fast charging process is optimized, the response sensitivity to the precursor signal of the thermal runaway risk is enhanced by combining the coupling analysis of the heat flux offset and the energy conversion ratio, the risk level is divided into intervals by the stage fluctuation intensity distribution characteristics, the accuracy of risk assessment is improved, the risk state is continuously monitored in the time sequence by combining the trend continuity and the multi-parameter matching intensity information, the operation safety and the assessment accuracy under the high-rate charging condition are enhanced, and the abnormal state judgment efficiency under the complex thermal-electricity imbalance condition is improved. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The system flowchart of the application is shown in the figure; Figure 2 The heat diffusion monitoring module flowchart of the application is shown in the figure; Figure 3 The voltage response identification module flowchart of the application is shown in the figure; Figure 4 The pressure difference trend extraction module flowchart of the application is shown in the figure; Figure 5 The energy offset determination module flowchart of the application is shown in the figure; Figure 6 The risk level output module flowchart of the application is shown in the figure. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical scheme and advantages of the application clearer, the application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the application and do not limit the application.

[0016] In the description of the application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the application and simplify the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the application. In addition, in the description of the application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0017] Please refer toFigure 1 The lithium battery thermal runaway risk warning system based on fast charging scenarios includes: The heat diffusion monitoring module calls the heat flux data on the surface of the battery cell, calculates the ratio difference of heat flux change in adjacent areas, and determines the direction of change in continuous cycles. Combining the trend of the ratio difference between the current cycle and the previous cycle, it identifies the location with the largest change and generates information on the degree of heat flux offset. The voltage response identification module extracts temperature rise and voltage change data in the corresponding region based on heat flux offset information, calculates the directional combination of the two, filters out reverse offset relationships, analyzes the synchronization degree of the combination sequence in combination with time interval, and generates thermoelectric response hysteresis amplitude. The differential pressure trend extraction module calls the thermoelectric response hysteresis amplitude to extract the differential pressure data of the positive and negative electrodes of the cell within the corresponding period, analyzes the difference in the direction of change and the growth rate, combines the stability of growth and direction, filters the same growth sequence, extracts the point with the maximum differential pressure amplitude, and generates differential pressure trend amplitude information. The energy offset determination module calls the differential pressure trend amplitude information, extracts the current and temperature rise rate data, judges the synchronicity between the ratio direction and the current change, combines the current change amplitude and the temperature rise rate, analyzes the matching degree of electrothermal energy response, and generates energy conversion offset information. The risk level output module calls the energy conversion offset information, extracts the direction of change of multiple parameters within the same period, calculates the matching strength of the temperature rise and the reverse change of voltage, combines the consistency of the pressure difference direction, calculates the change strength between the target trend and the energy conversion offset ratio, and outputs the risk level division interval information based on the phased fluctuation distribution characteristics of the strength in the sequence.

[0018] Information on heat flux deviation includes the range of the main heat diffusion path, the trend of heat flux ratio difference, and the location of abnormal heat flux concentration. The thermoelectric response hysteresis amplitude specifically refers to the directional combination relationship of voltage and temperature rise, the time series of directional deviation, and the characteristics of the degree of synchronization of change. The pressure difference trend amplitude information specifically includes the direction of pressure difference growth trend, the maximum pressure difference growth point, and the direction of temperature rise coupling trend. Energy conversion deviation information includes the direction of change of current and temperature rise ratio, the synchronization trend of electrothermal response, and the current sudden change segment. The risk level classification interval information specifically refers to the strength of reverse change matching, the stage of cross trend distribution, and the joint strength of electrothermal deviation.

[0019] Please see Figure 2 The thermal diffusion monitoring module includes: The heat flux acquisition submodule calls the heat flux data on the surface of the battery cell, analyzes the range of heat flux variation in multiple directional regions on the surface of the battery cell, calculates the ratio difference of heat flux variation between adjacent regions, and generates the heat flux variation interval. The system invokes a pre-defined database of heat flux on the cell surface, which stores data from nine heat flux sensors (arranged in a 3×3 matrix and numbered as follows) surrounding the cell surface. to In each monitoring cycle (cycle duration set to...) The real-time data collected by the system, firstly, in the current cycle In the middle, extract the central area (sensor) The heat flux data is as follows: And extract the heat flux data of its four orthogonally adjacent regions, specifically the upper region ( )of The lower area ( )of left side area ( )of and the right side area ( )of Then, the system uses the central area heat flux Using this as a benchmark, the ratio difference between the heat flux of the central region and that of each adjacent region is calculated. The calculation process involves subtracting the heat flux value of the central region from the heat flux value of the adjacent regions, and then dividing the difference by the heat flux value of the central region, thus obtaining four ratio differences. Specifically, the calculation is as follows: The ratio difference ,and The ratio difference ,and The ratio difference ,and The ratio difference Finally, the system performs set analysis on these calculated ratio differences to determine the minimum value in the set. and maximum value and the closed interval formed by these two values ​​[ , As a result, the range of heat flux variation is generated.

[0020] The ratio trend screening submodule calculates the ratio difference of heat flux change in adjacent regions in each cycle based on the heat flux change range, determines the direction of change of the ratio difference in continuous cycles, and filters the region sequence with the same trend by combining the ratio difference change trend of the current cycle and the previous cycle to obtain the ratio trend sequence. Based on the period Time Center Area Heat flux variation range [ , ], for three consecutive monitoring cycles ( The system processes the data within a given period. First, it calculates the central region for each cycle. With all its adjacent areas ( The difference in heat flux ratio, for example, during a cycle. At that time, the heat flux values ​​were respectively The calculated ratio difference sequence is [ ], in the cycle At that time, the heat flux values ​​were respectively , The calculated ratio difference sequence is [ Subsequently, the system determines the direction of change of the ratio difference between adjacent regions over a continuous period, in order to... Taking the region as an example, it is in The ratio differences within the three periods are respectively ,because and The trend shows a continuous decrease, therefore the trend is judged as "negative." The system applies the same trend judgment to all adjacent regions and compares it with... The region constitutes the region pair [ ],[ ],[ ],[ All were determined to be in a negative trend. Finally, the system selected all the regional sequences with consistent trends. Since the ratio difference of all four adjacent regions showed a continuously decreasing negative trend, all these regional sequences were retained to obtain the ratio trend sequence.

[0021] The main path determination submodule calls the ratio trend sequence, compares the ratio change amplitude of each region in the sequence, extracts the position with the largest amplitude as the range of the main heat diffusion path, and obtains information on the degree of heat flux offset. Call the containing region pair [ ],[ ],[ ],[ The ratio trend sequence of ], for each region within the sequence during the monitoring period (cycle) to To compare the rate changes within each region, the system first calculates the total rate change for each region's rate difference, which is defined as the final period ( ) and the initial period ( The absolute difference of the ratio difference, for the region Its range of change is -( ) = For the region Its range of change is -( ) = For the region Its range of change is -( ) = , for the region , the change range is -( ) = , then the system compares the four calculated change range values , determines the maximum value as , and the region corresponding to the value is , so the system determines the direction of heat diffusion from the central region to the right region as the main path of heat diffusion, and takes the identification result of the path range and the corresponding maximum change range as the core parameter to obtain the heat flux offset degree information.

[0022] Please refer to Figure 3 , the voltage response identification module includes: The rate extraction and analysis submodule extracts the cell voltage data and temperature rise change data of the corresponding region based on the heat flux offset degree information, analyzes the voltage increment value and temperature increment value in each period, obtains the voltage change rate and temperature rise rate data, and obtains the rate sequence processing result by combining the normalization processing; Based on the heat flux offset degree information, the information indicates that the main path range of heat diffusion is the region to , the system immediately locks this range and extracts the cell voltage data and temperature rise change data of the corresponding region in the problem diagnosis time period (period to ), specifically, the voltage value sequence and the temperature value sequence of the region are retrieved from the data record, the period length is , the system analyzes the voltage increment value and temperature increment value in each period, for period , the voltage increment , the temperature increment , for period , the voltage increment , the temperature increment , thereby obtaining the voltage change rate sequence and the temperature rise rate sequence , and then performing normalization processing, the setting of the normalization reference value is based on the statistical analysis of 1000 times of cycle test data of the battery under standard fast charging conditions, and the maximum value of the voltage change rate historical absolute value and the temperature rise rate historical maximum value As a benchmark, the current rate is divided by the benchmark value to obtain the period The normalized rate is , , the period The normalized rate is , , the rate sequence processing result is obtained.

[0023] The direction matching judgment submodule calls the rate sequence processing result, analyzes the directionality characteristics of the temperature rise rate and the voltage change rate in each period, calculates the combined response relationship of the two in each period, screens the combined relationship of the reverse offset, and obtains the time period and the corresponding index section formed by the continuous period of the reverse combination, extracts the combined sequence formed by the continuous offset behavior, and obtains the reverse offset section sequence; Call the rate sequence processing result in the period The rate sequence processing result in the period And , that is, the normalized voltage change rate sequence[ ] and the normalized temperature rise rate sequence[ ], the system analyzes the directionality characteristics of the temperature rise rate and the voltage change rate in each period one by one, the directionality is determined by the sign of the numerical value, the positive value represents increase or rise, and the negative value represents decrease or fall. In the period , the temperature rise rate is positive, and the voltage change rate is negative, and the directions of the two are opposite, forming a reverse offset combination. In the period , the temperature rise rate is positive, and the voltage change rate is negative, also forming a reverse offset combination. The system screens all the combined relationships of the reverse offset. Since the reverse offset appears continuously from the period To , the system determines this time period[ ] as a time period formed by a continuous reverse combination, and records its corresponding index section, that is, the period numbers 6 and 7. Based on this, the system extracts the offset behavior of the two continuous periods to form a combined sequence, that is, , to obtain the reverse offset section sequence.

[0024] The response lag calculation submodule calls the reverse offset section sequence, obtains the voltage change rate and the temperature rise rate value of each period point in the sequence, and uses the formula: ; Calculate the degree of asynchrony in the time and amplitude dimensions, and determine the overall lag response strength of the period section to obtain the thermoelectric response lag amplitude; Among them, is the degree of asynchronization in time and amplitude dimension, which is obtained by averaging the normalized rate difference of each cycle in the reverse offset combination sequence, is the normalized voltage rate of the th cycle, which is obtained by extracting the cell voltage data, calculating the voltage increment of adjacent sampling points divided by the cycle length to obtain the original voltage rate of change, and normalizing it based on the maximum rate in the reference cycle, is the normalized temperature rate of the th cycle, which is obtained by extracting the cell temperature data, calculating the temperature increment of adjacent sampling points divided by the cycle length to obtain the original temperature rate of change, and normalizing it based on the maximum rate in the reference cycle, is a positive constant to avoid zero denominator, is the number of consecutive cycles corresponding to the reverse offset combination, which is obtained by identifying the cycles with opposite directions of normalized voltage rate and temperature rate and counting their number, is the cycle index value, used to traverse each cycle in the reverse offset combination section; call the reverse offset section sequence[ ], and get the voltage rate of change normalized value and the temperature rate of change normalized value of each cycle point in the sequence from the rate sequence processing result, specifically cycle , and , and cycle , and , calculate the degree of asynchronization in time and amplitude dimension using the formula , in this example, the number of cycles summed is 2, representing the number of consecutive cycles corresponding to the reverse offset combination, i.e. from index value to , the constant is a positive number set to prevent zero denominator, its value is determined according to the hardware calculation accuracy and signal noise level, by evaluating the minimum value of in historical data, set to , which is a value much smaller than the normal order of magnitude of the calculation result, the calculation process first expands the summation part, calculates the terms of and respectively, when , the value of the term is , when , the value of the term is , then add the values of all terms and divide by , i.e. ; The degree of asynchrony in the time and amplitude dimensions refers to whether there is a difference in response time (time dimension) and consistency in amplitude between the voltage change rate and temperature rise rate per unit time during cell operation. When voltage response fluctuates while temperature remains relatively stable, or when temperature rises sharply while voltage changes slowly, asynchrony is observed, indicating a break in the response chain or abnormal structural coupling in the thermal and electrical behavior. A larger parameter value indicates stronger asynchrony, reflecting potential issues such as energy release path deviation, enhanced electrochemical polarization, or delayed interlayer heat transfer between the cell's internal heat source and electric field. This serves as a crucial indicator for early warning of impending thermal runaway. The logic of the formula lies in the molecule... The difference between the normalized voltage and temperature rates was directly calculated, reflecting the inconsistency in their magnitude of change. (The denominator...) As a normalization factor for geometric mean, it is used to adjust for the influence of the molecular difference on the magnitude of the rate itself. When the absolute values ​​of both rates are large or small, their product will change accordingly, thus making... The absolute value more stably reflects relative asynchrony rather than absolute rate difference. Absolute value calculation ensures the degree of asynchrony is positive, and the final averaging yields the overall hysteresis response strength for the entire segment. This refers to the magnitude of the thermoelectric response hysteresis. The results indicate that within the identified reverse offset segment, there is a significant asynchrony between the voltage change rate and the temperature rise rate in both amplitude and time response. A larger value indicates a greater deviation from the coordinated state. The formula standardizes the difference by introducing the square root of the rate product, effectively measuring the relative mismatch between voltage and temperature responses during dynamic changes, rather than simply comparing their absolute differences. This design makes the evaluation results more robust to rate fluctuations of varying magnitudes, thus enabling more accurate identification of thermoelectric coupling failures caused by internal electrochemical anomalies.

[0025] Please see Figure 4 The differential pressure trend extraction module includes: The differential pressure change analysis submodule calls the thermoelectric response hysteresis amplitude to extract the differential pressure data between the positive and negative electrodes of the cell for the corresponding period, analyzes the direction of differential pressure change in continuous periods, and generates a sequence of differential pressure change directions. Calling thermoelectric response hysteresis amplitude This value exceeds the preset differential pressure analysis trigger threshold. This threshold was determined by analyzing 50 sets of battery experimental data ranging from normal to thermal runaway, identifying that the thermoelectric response hysteresis amplitude generally jumps to a certain level near the point of thermal runaway. Based on the above settings, the system then extracts the core period in which the thermoelectric response hysteresis occurs. ) and its subsequent cycles ( The data sequence is as follows: [The data includes the voltage difference between the positive and negative electrodes of the battery cell.] Next, the system analyzes the direction of pressure difference change in this continuous periodic sequence, which is achieved by calculating the pressure difference between adjacent periods. arrive The difference is ,cycle arrive The difference is ,cycle arrive The difference is Since all calculated differences are positive, it indicates that the pressure difference is increasing in each cycle. The system records the direction of change in each cycle as "increase" and generates a sequence of pressure difference change directions.

[0026] The growth continuation screening submodule calculates the difference in the growth rate of pressure difference between adjacent periods based on the pressure difference change direction sequence, judges the consistency of the change direction and locates the trend continuation segment, filters the periodic sequence of continuous same-direction growth, and obtains the same-direction growth segment data. Based on a pressure differential change direction sequence where all cycles show an "increase," the difference in the magnitude of pressure differential increase between adjacent cycles is calculated to determine whether the growth trend is continuing or strengthening. First, the system obtains the growth magnitude for each week, as follows: Then, calculate the differences between these amplitudes. arrive amplitude and arrive The difference in amplitude is , arrive amplitude and arrive The difference in amplitude is Since these differences are all positive, it indicates that the rate of increase in pressure differential is continuously accelerating, and the direction of change is clearly consistent. Based on this, the system can determine the starting point of the cycle. arrive The entire segment is considered a trend continuation segment, and this segment, composed of periodic sequences, is selected. The data of the same-direction growth segment is obtained from the continuous cyclical sequence of the same-direction growth.

[0027] The trend amplitude determination submodule compares the differential pressure change amplitude of all cycles within the same growth segment based on the data of the same growth segment, extracts the cycle point with the largest differential pressure change amplitude, and combines it with the temperature rise change direction data corresponding to the cycle to obtain differential pressure trend amplitude information. Based on the data of the same growth segment, i.e., the periodic sequence [ ] and the corresponding pressure difference increase[ ], directly compare the amplitude of the pressure difference change of all periods in the section, and the maximum value is ] in the numerical set , which occurs in the transition from period to , so the system extracts the period point with the maximum amplitude of the pressure difference change as period , then the system calls the temperature rise change direction data corresponding to period , at this time the temperature rise rate is , which is positive, representing a continuous temperature rise, the system binds the period point with the maximum amplitude of the pressure difference change ( ) and the temperature rise continuous (positive) information at this point to obtain the pressure difference trend amplitude information.

[0028] Please refer to Figure 5 , the energy offset determination module includes: The ratio direction calculation submodule extracts the current change rate and temperature rise change rate data in the corresponding period based on the pressure difference trend amplitude information, analyzes the ratio change direction of the current change rate and the temperature rise change rate in consecutive periods, and establishes a ratio direction sequence; Based on the pressure difference trend amplitude information, which locks the key period point and the trend section[ ] before it, the system extracts the current change rate data and the temperature rise change rate data of the corresponding period in this section, assuming that the current change rate sequence is[ ] and the temperature rise change rate sequence is[ ], then the system analyzes the ratio of the current change rate and the temperature rise change rate in consecutive periods, in period , the ratio is , in period , the ratio is , in period , the ratio is , in period , the ratio is , then the system determines the change direction of the ratio by comparing the ratio values of adjacent periods, , , , so from to , the change direction of the ratio is "up", and the system establishes a ratio direction sequence with all "up" contents accordingly.

[0029] The synchronous trend filtering submodule calls the ratio directional sequence to extract the current change direction information within the corresponding period, calculates the change amplitude of the ratio between adjacent periods, determines the synchronicity between the ratio change direction and the current change direction, filters the synchronously rising period sequence in continuous periods, and generates a synchronously rising trend section sequence by combining the temperature rise rate data within the section. Call the ratio directionality sequence, which is displayed in the period. to As the internal ratio continues to "increase," the system further extracts information on the direction of current change within that segment, due to the current change rate sequence [ All values ​​in the [data] are positive, indicating that the current itself is also trending upwards. The system determines that the direction of the ratio change ("rising") and the direction of the current change ("rising") are synchronous. Therefore, it selects values ​​from the period [data]. arrive The entire sequence is taken as a synchronously rising periodic sequence within a continuous period. Subsequently, the system compares this periodic sequence with the temperature rise rate data within that segment. The data are integrated to generate a sequence of synchronously rising trend segments.

[0030] The response matching and identification submodule extracts the current surge amplitude and temperature rise rate within the target segment based on the synchronous upward trend segment sequence. Combined with the differential pressure change rate, it analyzes the adjustment offset strength between the current and temperature rise rates within the target period, using the following formula: ; Calculate the electrothermal response matching offset value, and combine the concentrated distribution characteristics of the offset intensity in each cycle to obtain energy conversion offset information; in, This is the electrothermal response matching offset value, representing the degree of difference in energy response between changes in current and temperature rise within the target period. For the first The normalized value of the current change rate within each cycle is obtained by processing the original current change rate data using a min-max normalization method. For the first The normalized value of the rate of temperature change within each cycle is obtained by normalizing the rate of temperature change. For the first The normalized value of the rate of change of voltage difference within each cycle is obtained by extracting the continuous cycle rate of change based on the positive and negative electrode voltage difference data and then normalizing it. This is the index number of the current processing cycle in the continuous cycle sequence, with a value range of... to , The total number of consecutive periods indicates the number of periods involved in a synchronous upward trend segment. The normalized value of the rate of change of the current in the target period is calculated by normalizing the rate of change of the current data in the period, The normalized value of the rate of change of the temperature rise in the target period , The number position of the target period in the trend sequence, indicating the specific period corresponding to the matching offset value; According to the synchronous rising trend segment sequence[ ], the current mutation amplitude and the rate of change of the temperature rise in the segment are extracted, and the adjustment offset strength between the current and the rate of change of the temperature rise in the target period is analyzed in combination with the differential pressure change rate. The system uses the formula to calculate, where is the total number of periods in the segment, which is 4 here, is the period index, from 6 to 9, the target period is set to the last period of the segment, i.e. , first normalize the required data, based on historical normal operation data, set the maximum values of current, temperature rise, and differential pressure change rate as , , , the data processing results are shown in Table 1.

[0031] Table 1. Parameters for calculating the matching of the electro-thermal response

[0032] As shown in Table 1, the original and normalized data required for calculating the matching offset value of the electro-thermal response are listed. The calculation logic of the formula is that the sum part calculates the average value of the weighted "current-temperature rise" response ratio in the entire segment considering the influence of the differential pressure, where serves as a weight item, which decreases when the differential pressure change rate increases, reducing the weight of the corresponding period response ratio in the average value, which reflects the disturbance of the internal pressure surge to the normal energy transmission path. The outside the sum item is the "current-temperature rise" response ratio of the target period, and the absolute value of the difference between the two is obtained, which is the deviation of the target period response behavior from the average behavior of the segment, i.e. The calculation process is as follows: first, calculate each term in the sum. For , the term value is , for , the term value is , for , the term value is , the average value of the sum is (Note that the sum item is only up to m-1, i.e. j=6,7,8), the response ratio of the target period is , then: ; wherein, the electric-thermal response matching offset value refers to the deviation intensity of the response coupling degree between the current change rate and the temperature rise change rate in the current target period compared to the overall trend sequence, and the value also considers the interference of the pressure difference change rate on the energy conduction path. When the value is close to zero, it means that the electric-thermal response behavior of the target period matches the overall trend to a high degree, and the energy conversion process is relatively coordinated; on the contrary, the greater the offset value, the more obvious the mismatch between the current input and the heat release in the period, which may indicate energy retention or abnormal conduction in the battery cell, and has strong precursor characteristics of thermal runaway. Therefore, this parameter is used in the system to locate the key period stage where thermal stability is destroyed, and is an important basic index for hierarchical risk warning. Finally, the system obtains the energy conversion offset information by combining the concentration distribution characteristics of the offset intensity in each period. The formula not only compares the rate relationship between the current input and the heat generation, but also innovatively introduces the pressure difference change rate as an adjustment factor, making the judgment of energy conversion imbalance more accurate and able to identify more hidden thermal runaway precursors caused by rapid internal pressure changes.

[0033] Referring to Figure 6 , the risk level output module comprises: The parameter combination extraction submodule calls the energy conversion offset information, extracts the temperature rise change rate, voltage change rate and pressure difference change direction data in the same period, analyzes the change direction of each parameter in the period sequence, and generates a parameter direction combination sequence; The energy conversion offset information is called, which contains a key electric-thermal response matching offset value , and the value is associated with the key period . The system then extracts the temperature rise change rate, voltage change rate and pressure difference change direction data in the same period , obtains the temperature rise change rate of period , , the voltage change rate , and the pressure difference change direction from period to is "growth" (based on the increment of ), and the system then analyzes the change direction combination of the three parameters in the period sequence, encodes the positive and negative signs of the rates and the classification of the pressure difference direction (growth / decline), for example, "positive" is encoded as 1, "negative" is encoded as -1, and "growth" is encoded as 1. Therefore, the combination in period is [ ], and the system also performs the same operation on the previous periods (such as ) to form a time sequence of parameter direction combinations and generate a parameter direction combination sequence.

[0034] The reverse matching judgment submodule calculates the matching strength of the reverse change of temperature rise and voltage based on the parameter direction combination sequence, combines the time consistency of the matching strength and the pressure difference direction continuation feature, judges the distribution feature of the multi-parameter cross trend in the time sequence, and obtains the multi-parameter trend distribution information; Based on the parameter direction combination sequence, the sequence is analyzed, the matching strength of the reverse change of temperature rise and voltage is calculated, and the calculation method of the matching strength is: in a time window (for example, from to ), the number of periods in which the temperature rise direction is “positive” and the voltage direction is “negative” is counted, and then divided by the total number of periods in the window. In this example, the temperature rise direction of period is positive, and the voltage direction is negative, so the matching strength is , which is a very high matching strength. Then, the system combines the matching strength and the time consistency of the pressure difference direction continuation feature in the same time window , and the continuation feature of the pressure difference direction is continuous “growth”, which is completely consistent in time with the continuous reverse change of temperature rise and voltage. The system judges that the cross trend of multi-parameters (temperature rise, voltage, pressure difference) presents a highly concentrated distribution feature in the time sequence, and obtains the multi-parameter trend distribution information.

[0035] The grade interval output submodule calculates the change strength between the target trend and the energy conversion offset ratio according to the multi-parameter trend distribution information, and outputs the risk grade division interval information according to the stage fluctuation distribution feature of the strength in the sequence; According to the multi-parameter trend distribution information, it is confirmed that there is a highly consistent dangerous feature set from period to , the system further calculates the change strength between the target trend and the energy conversion offset ratio, the change strength is defined as the product of the multi-parameter trend matching strength and the maximum energy conversion offset value, that is, , and the strength value is formed into a change strength sequence according to the time sequence to , assuming that the value is , the system divides the sequence into stages based on adjacent extreme turning points. Since the sequence is monotonically increasing, the entire sequence forms a stage. The stage fluctuation range is , the stage average strength is , and the stage continuous rise proportion is . The threshold is set based on the statistical analysis of the data set recorded in the abuse test of 200 different aging degree batteries. The strength threshold is set to the median of the data set , the fluctuation threshold is set as the median of the fluctuation range , the continuity threshold is set as the median of the proportion , the system matches the calculation result of the current stage[ ] with the threshold[ ], since all the indexes of the current stage exceed the corresponding threshold, the preset combination relationship of the highest risk level is met, and the risk level division interval information is output as "high risk".

[0036] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can use the disclosed technical content to make changes or modifications into equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made to the above embodiments without departing from the technical solution content of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. A lithium battery thermal runaway risk early warning system based on fast charging scenarios, characterized in that, The system comprises: The heat diffusion monitoring module calls the cell surface heat flux data, calculates the ratio difference of the heat flux change of adjacent regions, and judges the change direction in the continuous period, combines the trend of the ratio difference of the current and the last period, identifies the position with the largest change amplitude, and generates heat flux offset degree information; The voltage response identification module extracts the corresponding region temperature rise and voltage change data based on the heat flux offset degree information, calculates the directionality combination of the two, screens the reverse offset relationship, analyzes the synchronization degree of the combination sequence in combination with the time interval, and generates the heat-electricity response lag amplitude; The differential pressure trend extraction module calls the heat-electricity response lag amplitude, extracts the positive and negative electrode differential pressure data in the corresponding period, analyzes the change direction and growth amplitude difference, combines the stability of growth and direction, screens the same direction growth sequence, extracts the maximum point of differential pressure amplitude, and generates differential pressure trend amplitude information; The energy offset determination module calls the differential pressure trend amplitude information, extracts the current and temperature rise change rate data, judges the synchronization of the ratio direction and current change, analyzes the electric-thermal energy response matching degree in combination with the current mutation amplitude and temperature rise change rate, and generates energy conversion offset information. 2.The lithium battery thermal runaway risk pre-warning system based on fast charging scenario according to claim 1, wherein, The heat flux offset degree information includes heat diffusion main path range, heat flux ratio difference change trend, and heat flux abnormal concentration position, the heat-electricity response lag amplitude specifically refers to voltage temperature rise directionality combination relationship, directionality offset time sequence, and change synchronization degree feature, the differential pressure trend amplitude information specifically refers to differential pressure growth trend direction, maximum growth point of differential pressure, and temperature rise coupling trend direction, and the energy conversion offset information includes current and temperature rise ratio change direction, electric-thermal response synchronization trend, and current mutation section. 3.The fast-charging scenario based lithium battery thermal runaway risk pre-warning system of claim 1, wherein, The heat diffusion monitoring module comprises: The heat flux collection sub-module calls the cell surface heat flux data, analyzes the heat flux change range in multiple direction regions of the cell surface, calculates the ratio difference of the heat flux change between adjacent regions, and generates a heat flux change interval; The ratio trend screening sub-module calculates the ratio difference of the heat flux change of adjacent regions in each period based on the heat flux change interval, judges the change direction of the ratio difference in the continuous period, combines the ratio difference change trend of the current period and the last period, screens the region sequence with consistent trend, and obtains a ratio trend sequence; The main path determination sub-module calls the ratio trend sequence, compares the ratio change amplitudes of each region in the sequence, extracts the position with the largest amplitude as the heat diffusion main path range, and obtains the heat flux offset degree information.

4. The lithium battery thermal runaway risk early warning system based on fast charging scenarios according to claim 3, characterized in that, The voltage response identification module comprises: The rate extraction and analysis sub-module extracts the cell voltage data and temperature rise change data of the corresponding region based on the heat flux offset degree information, analyzes the voltage increment value and temperature increment value in each period, obtains voltage change rate and temperature rise rate data, combines normalization processing, and obtains rate sequence processing result; The direction matching judgment submodule calls the rate sequence processing result, analyzes the directionality characteristics of the temperature rise rate and the voltage change rate in each period, calculates the combined response relationship of both in each period, screens the combined relationship of reverse offset, and obtains the time period and the corresponding index section formed by the reverse combination in the continuous period, extracts the combined sequence formed by the continuous offset behavior, and obtains the reverse offset section sequence; The response lag calculation submodule calls the reverse offset section sequence, obtains the voltage change rate and the temperature rise rate value of each period point in the sequence, uses the formula: ； The calculation time and the degree of asynchrony in the amplitude dimension, and judges the overall lag response strength of the period section, and obtains the thermal-electric response lag amplitude value; wherein, is a degree of non-synchronization of time and amplitude dimensions, is a voltage change rate normalized value of the is a voltage change rate normalized value of the is a temperature rise rate normalized value of the is a temperature rise rate normalized value of the is a positive constant for avoiding zero denominator, is a number of consecutive periods corresponding to the reverse offset combination, is a period index value.

5. The lithium battery thermal runaway risk pre-warning system based on fast charging scenario according to claim 4, characterized in that, The differential pressure trend extraction module comprises: The differential pressure change analysis submodule calls the thermal-electric response lag amplitude value, extracts the differential pressure data of the positive electrode and the negative electrode of the battery cell corresponding to the period, analyzes the change direction of the differential pressure in the continuous period, and generates a differential pressure change direction sequence; The growth continuation screening submodule calculates the difference of the differential pressure growth amplitude of adjacent periods based on the differential pressure change direction sequence, judges the consistency of the change direction and locates the trend continuation section, screens the period sequence of continuous same-direction growth, and obtains same-direction growth section data; The trend amplitude determination submodule compares the differential pressure change amplitudes of all periods in the section according to the same-direction growth section data, extracts the period point with the maximum differential pressure change amplitude, and combines the temperature rise change direction data corresponding to the period to obtain the differential pressure trend amplitude information. 6.The fast-charging scenario based lithium battery thermal runaway risk pre-warning system according to claim 5, characterized in that, The energy offset determination module comprises: The ratio direction calculation submodule extracts the current change rate and temperature rise change rate data in the corresponding period based on the differential pressure trend amplitude information, analyzes the ratio change direction of the current change rate and the temperature rise change rate in the continuous period, and establishes a ratio directionality sequence; The synchronous trend screening submodule calls the ratio directionality sequence, extracts the current change direction information in the corresponding period, calculates the change amplitude of the ratio of adjacent periods, judges the synchronism of the ratio change direction and the current change direction, screens the period sequence of continuous periods in which the ratio changes synchronously, and combines the temperature rise change rate data in the section to generate a synchronous upward trend section sequence; The response matching recognition submodule extracts the current mutation amplitude and the temperature rise change rate in the target section according to the synchronous upward trend section sequence, combines the differential pressure change rate, analyzes the adjustment offset strength between the current and the temperature rise change rate in the target period, calculates the electro-thermal response matching offset value, combines the centralized distribution characteristics of the offset strength in each period, and obtains the energy conversion offset information.

7. The fast-charging scenario based lithium battery thermal runaway risk early warning system according to claim 1, wherein, The system further comprises: The risk level output module calls the energy conversion offset information, extracts the change direction of multiple parameters in the same period, calculates the matching strength of the reverse change of the temperature rise and the voltage, combines the consistency of the differential pressure direction, calculates the change strength between the target trend and the energy conversion offset ratio, and outputs the risk level division interval information according to the phased fluctuation distribution characteristics of the strength in the sequence; The risk level division interval information specifically refers to the reverse change matching strength, the cross trend distribution stage, and the electro-thermal offset joint strength. 8.The fast-charging scenario based lithium battery thermal runaway risk pre-warning system of claim 7, wherein, The risk level output module comprises: The parameter combination extraction submodule calls the energy conversion offset information, extracts the temperature rise change rate, voltage change rate and pressure difference change direction data in the same period, analyzes the change direction of each parameter in the period sequence, and generates a parameter direction combination sequence; The reverse matching judgment submodule calculates the matching strength of the reverse change of temperature rise and voltage based on the parameter direction combination sequence, combines the time consistency of the matching strength and the pressure difference direction continuation characteristic, judges the distribution characteristics of the multi-parameter cross trend in the time sequence, and obtains multi-parameter trend distribution information; The grade interval output submodule calculates the change strength between the target trend and the energy conversion offset ratio according to the multi-parameter trend distribution information, and outputs risk grade division interval information according to the stage fluctuation distribution characteristics of the strength in the sequence. 9.The fast-charging scenario based lithium battery thermal runaway risk pre-warning system of claim 8, wherein, The process of outputting risk grade division interval information according to the stage fluctuation distribution characteristics of the strength in the sequence is specifically: forming a change strength sequence in time sequence according to the change strength, performing stage division on the change strength sequence based on adjacent extreme inflection points, and obtaining a stage set; For each stage, calculate the stage fluctuation range, the stage average strength, and the stage continuous rising proportion; Based on the median of the stage fluctuation range of all stages in the same change strength sequence, set the fluctuation threshold, based on the median of the stage average strength of all stages, set the strength threshold, and based on the median of the stage continuous rising proportion of all stages, set the continuity threshold; According to the fluctuation threshold, the strength threshold and the continuity threshold, the stage set is matched, and the risk grade division interval information meeting the threshold combination relationship is output.

Images