Battery thermal runaway multi-feature fusion early warning method, device, equipment and storage medium

Abstract

The application discloses a battery thermal runaway multi-feature fusion early warning method and device, equipment and a storage medium, and relates to the technical field of battery safety. The method comprises the following steps: obtaining a thermal runaway feature based on temperature channel data and environmental temperature; adjusting a dynamic threshold value and a weight corresponding to the thermal runaway feature based on the thermal runaway feature, to obtain an adjusted dynamic threshold value and an adjusted weight; determining a thermal runaway risk level based on the thermal runaway feature, the adjusted dynamic threshold value and the adjusted weight; and executing a corresponding hierarchical response control strategy according to the thermal runaway risk level. Since the technical means of dynamically updating feature threshold values and fusion weights based on real-time operation data is adopted, the technical problems of inaccurate early warning sensitivity and high false alarm rate of the existing fixed threshold method under battery aging, environmental changes and dynamic working conditions are solved, and earlier and more accurate identification of thermal runaway risks is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery safety, in particular to a battery thermal runaway multi-feature fusion early warning method and device, equipment and a storage medium. BACKGROUND

[0002] In a battery management system, early and accurate early warning of thermal runaway risk is needed to provide sufficient buffer time within tens of seconds to hundreds of seconds before thermal runaway occurs, and to implement progressive and reversible control strategies according to different risk levels, so as to maintain the usability of the system as much as possible under the premise of safety.

[0003] The current common implementation scheme mainly relies on setting fixed thresholds for battery absolute temperature and first-order temperature rise rate for judgment. This scheme cannot effectively distinguish between uniform temperature rise caused by normal high-rate charging and discharging and local thermal abnormalities caused by internal short circuits and other faults, and the criterion threshold cannot be self-adaptively adjusted according to long-term factors such as battery aging and environmental temperature changes.

[0004] Therefore, the prior art has the problem of high false alarm rate in dynamic conditions due to the dependence on fixed thresholds and single features.

[0005] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0006] The main purpose of the present application is to provide a battery thermal runaway multi-feature fusion early warning method, device, equipment and storage medium, aiming to solve the technical problem of how to improve the timeliness and accuracy of battery thermal runaway early warning in dynamic conditions.

[0007] To achieve the above purpose, the present application provides a battery thermal runaway multi-feature fusion early warning method, which comprises:

[0008] Obtaining temperature channel data, environmental temperature and adjustment weight in a battery system;

[0009] Based on the temperature channel data and the environmental temperature, a thermal runaway feature is obtained;

[0010] Based on the thermal runaway feature, the dynamic threshold corresponding to the thermal runaway feature is adjusted to obtain an adjusted dynamic threshold;

[0011] Based on the thermal runaway feature, the adjusted dynamic threshold and the adjustment weight, a thermal runaway risk level is determined;

[0012] According to the thermal runaway risk level, a corresponding graded response control strategy is executed.

[0013] In one embodiment, obtaining thermal runaway characteristics based on the temperature channel data and the ambient temperature includes:

[0014] Based on the temperature channel data, hotspot temperature, temperature gradient characteristics, window average temperature, periodic average temperature, and multiple first-order heating rates are obtained.

[0015] Obtain the heating rate and heat dissipation model under static operating conditions;

[0016] Based on the heating rate under static conditions, the heat dissipation model, the periodic average temperature, and the ambient temperature, the self-heating residual of the static section is obtained.

[0017] The average window temperature, the first-order heating rate, the hot spot temperature, the temperature gradient characteristics, and the self-heating residual of the static section are used as thermal runaway characteristics.

[0018] In one embodiment, adjusting the dynamic threshold corresponding to the thermal runaway characteristic based on the thermal runaway characteristic to obtain the adjusted dynamic threshold includes:

[0019] Obtain the exponential forgetting coefficient, historical exponential forgetting mean, and historical forgetting variance corresponding to the thermal runaway characteristics;

[0020] When the thermal runaway features do not meet the anomaly removal criteria, the current exponential forgetting mean is obtained based on the historical exponential forgetting mean, the exponential forgetting coefficient, and the thermal runaway features.

[0021] Based on the thermal runaway characteristics, the historical exponential forgetting mean, the exponential forgetting coefficient, and the historical forgetting variance, the current forgetting variance is obtained;

[0022] Based on the current exponential forgetting mean and the current forgetting variance, the dynamic threshold corresponding to the thermal runaway feature is adjusted to obtain the adjusted dynamic threshold.

[0023] In one embodiment, adjusting the dynamic threshold corresponding to the thermal runaway feature based on the current exponential forgetting mean and the current forgetting variance to obtain the adjusted dynamic threshold includes:

[0024] Obtain operating condition bias, high-risk classification coefficient, and critical classification coefficient;

[0025] The high-risk classification threshold is obtained based on the current index forgetting mean, the current forgetting variance, the operating condition bias, and the high-risk classification coefficient.

[0026] The critical grading threshold is obtained based on the current index forgetting mean, the current forgetting variance, the operating condition bias, and the critical grading coefficient.

[0027] The high-risk classification threshold and the critical classification threshold are used as the dynamic adjustment thresholds.

[0028] In one embodiment, determining the thermal runaway risk level based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight includes:

[0029] Based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight, a characteristic risk score is obtained;

[0030] Based on the aforementioned feature risk scores, a comprehensive risk score is obtained;

[0031] Obtain the reference bias, current-dependent slope, and bus current;

[0032] Based on the aforementioned reference bias, current-related slope, and bus current, a comprehensive risk classification threshold is obtained;

[0033] Based on the thermal runaway characteristics, the adjusted dynamic threshold, and the comprehensive risk classification threshold, the classification conditions are determined.

[0034] Based on the comprehensive risk score, the thermal runaway characteristics, and the grading conditions, the thermal runaway risk level is determined.

[0035] In one embodiment, executing a corresponding graded response control strategy based on the thermal runaway risk level includes:

[0036] When the thermal runaway risk level is the first risk level, the strategy is to suspend the active battery balancing operation and record the enhanced log.

[0037] When the thermal runaway risk level is the second risk level, based on the response control strategy of the first risk level, a strategy of limiting the battery charging and discharging power and activating or strengthening the heat dissipation system is added.

[0038] When the thermal runaway risk level is the third risk level, a strategy is implemented to disconnect the battery main circuit, maintain the heat dissipation system at maximum operation, and write critical data into non-volatile memory. The first risk level is lower than the second risk level, and the second risk level is lower than the third risk level.

[0039] In one embodiment, based on the thermal runaway characteristics, after executing the corresponding graded response control strategy according to the thermal runaway risk level, the process includes:

[0040] The contribution of events with a thermal runaway risk level of the second or third risk level is statistically analyzed.

[0041] The adjustment weights are updated based on the contribution level to obtain the updated adjustment weights.

[0042] In addition, to achieve the above objectives, this application also proposes a battery thermal runaway multi-feature fusion early warning device, which includes: battery thermal runaway multi-feature fusion early warning.

[0043] In addition, to achieve the above objectives, this application also proposes a battery thermal runaway multi-feature fusion early warning device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the battery thermal runaway multi-feature fusion early warning method as described above.

[0044] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the battery thermal runaway multi-feature fusion early warning method described above.

[0045] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the battery thermal runaway multi-feature fusion early warning method described above.

[0046] This application acquires temperature channel data, ambient temperature, and adjustment weights from the battery system; based on the temperature channel data and ambient temperature, it obtains thermal runaway characteristics; based on the thermal runaway characteristics, it adjusts the dynamic thresholds corresponding to the thermal runaway characteristics to obtain adjusted dynamic thresholds; based on the thermal runaway characteristics, the adjusted dynamic thresholds, and the adjustment weights, it determines the thermal runaway risk level; and according to the thermal runaway risk level, it executes a corresponding graded response control strategy. By employing a technique that dynamically updates the characteristic thresholds and fusion weights based on real-time operating data, it solves the technical problems of inaccurate warning sensitivity and high false alarm rates in existing fixed threshold methods under battery aging, environmental changes, and dynamic operating conditions. This achieves earlier and more accurate identification of thermal runaway risks and supports a progressive safety response from warning to shutdown, thereby improving safety while ensuring system availability. Attached Figure Description

[0047] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating an embodiment of the multi-feature fusion early warning method for battery thermal runaway in this application.

[0050] Figure 2 This is a schematic diagram of a hierarchical response state machine provided in Embodiment 1 of the battery thermal runaway multi-feature fusion early warning method of this application;

[0051] Figure 3 This is a system overall structure diagram provided in Embodiment 1 of the battery thermal runaway multi-feature fusion early warning method of this application;

[0052] Figure 4 This is a schematic diagram of the module structure of the battery thermal runaway multi-feature fusion early warning device according to an embodiment of this application;

[0053] Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the battery thermal runaway multi-feature fusion early warning method in the embodiments of this application.

[0054] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0055] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0056] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0057] The main solution of this application embodiment is as follows: acquire temperature channel data, ambient temperature, and adjustment weights in the battery system; obtain thermal runaway characteristics based on the temperature channel data and the ambient temperature; adjust the dynamic threshold corresponding to the thermal runaway characteristics based on the thermal runaway characteristics to obtain an adjustment dynamic threshold; determine the thermal runaway risk level based on the thermal runaway characteristics, the adjustment dynamic threshold, and the adjustment weights; and execute a corresponding graded response control strategy according to the thermal runaway risk level.

[0058] In this embodiment, for ease of description, the following description uses a computer as the execution subject.

[0059] Current technologies primarily rely on setting fixed thresholds for battery absolute temperature and first-order temperature rise rate. This approach cannot effectively distinguish between uniform temperature rise caused by normal high-rate charging and discharging and localized thermal anomalies caused by faults such as internal short circuits. Furthermore, its threshold criteria cannot be adaptively adjusted to long-term factors such as battery aging and changes in ambient temperature.

[0060] This application provides a solution that uses a technique based on real-time operational data to dynamically update feature thresholds and fusion weights. This solves the technical problems of inaccurate early warning sensitivity and high false alarm rate of existing fixed threshold methods under battery aging, environmental changes and dynamic operating conditions. It enables earlier and more accurate identification of thermal runaway risks and supports a progressive safety response from early warning to cutoff, thereby improving safety while ensuring system availability.

[0061] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or computer capable of performing the above functions. The following description uses a computer as an example to illustrate this embodiment and the subsequent embodiments.

[0062] Based on this, embodiments of this application provide a multi-feature fusion early warning method for battery thermal runaway, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the multi-feature fusion early warning method for battery thermal runaway according to this application.

[0063] In this embodiment, the battery thermal runaway multi-feature fusion early warning method includes steps S10~S50:

[0064] Step S10: Obtain temperature channel data, ambient temperature, and adjustment weights in the battery system;

[0065] It should be noted that a battery system refers to a power battery pack or energy storage battery unit that includes multiple battery cells, battery modules, and their associated management components connected in series or parallel. Temperature channel data refers to multiple instantaneous temperatures of different temperature channels collected by temperature sensors at preset intervals within a preset window time. Ambient temperature refers to a reference value reflecting the temperature of the external cooling medium of the battery system, measured by an independent temperature sensor installed on the outside of the battery box or at the system air inlet. Adjustment weights are coefficients used to characterize the relative importance of each thermal runaway feature when calculating the overall risk. For example, the initial value of each thermal runaway feature threshold, the high-risk weight of temperature rise rate, the critical weight of temperature rise rate, the high-risk weight of temperature difference, the critical weight of temperature difference, and the noise suppression coefficient.

[0066] For example, the temperature sensor may be a negative temperature coefficient thermistor, with a preset sampling period of 1 second and a preset window time of 60 seconds, thereby forming a sliding time window containing 60 sampling points for subsequent feature calculation.

[0067] It is understandable that since battery thermal runaway often begins with an abnormal temperature rise in a local cell and spreads throughout the system through thermal diffusion, step S10 can acquire comprehensive temperature information reflecting the overall and local, internal and external thermal states. This lays a data foundation for accurately identifying early thermal runaway characteristics and avoids missed or delayed warnings due to a single measurement point or incomplete information, thereby improving the system's ability to perceive potential thermal runaway risks.

[0068] Step S20: Based on the temperature channel data and the ambient temperature, obtain the thermal runaway characteristics;

[0069] It should be noted that thermal runaway characteristics are a series of dynamic parameters used to quantitatively characterize the evolution trend of a battery system toward a thermal runaway state.

[0070] Understandably, since thermal runaway is a gradual process driven by the coupling of multiple physical and chemical processes, a single absolute temperature value cannot fully reflect its early signs. Therefore, step S20 can extract multi-dimensional and more representative dynamic features from the original temperature data, avoiding inaccurate warnings caused by relying on a single criterion, thereby improving the comprehensiveness and reliability of risk identification.

[0071] In one feasible implementation, step S20 may include: obtaining hot spot temperature, temperature gradient characteristics, window average temperature, periodic average temperature, and multiple first-order heating rates based on the temperature channel data; obtaining the quiescent heating rate and heat dissipation model; obtaining the quiescent section self-heating residual based on the quiescent heating rate, the heat dissipation model, the periodic average temperature, and the ambient temperature; and using the window average temperature, the first-order heating rate, the hot spot temperature, the temperature gradient characteristics, and the quiescent section self-heating residual as thermal runaway characteristics.

[0072] It should be noted that: hotspot temperature refers to the highest temperature value identified among all current temperature channel data, used to locate the hottest point; temperature gradient feature is an indicator used to quantify the uniformity of temperature distribution within the battery system, which can be obtained by calculating the difference or dispersion between temperature sensor readings at different locations; window average temperature refers to the arithmetic mean of the instantaneous temperature values ​​of all effective temperature channels within a preset sliding time window; periodic average temperature refers to the average of the instantaneous readings of all temperature channels within a single sampling period; first-order heating rate refers to the change in temperature per unit time, calculated by comparing the temperature values ​​of the same temperature channel in the current period with those in the previous period; quiescent heating rate refers to the heating rate calculated based on temperature channel data during the period when the system is determined to be in a quiescent condition; the heat dissipation model is a functional model used to estimate the intensity of heat dissipation from the battery system to the environment under given conditions; quiescent self-heating residual refers to the net heating rate obtained by subtracting the theoretical heat dissipation rate estimated by the heat dissipation model from the measured quiescent heating rate, used to indicate the heat generated by side reactions inside the battery.

[0073] For example, the temperature gradient characteristic can be specifically reflected as the difference between the readings of the highest and lowest temperature sensors within the same module; the heat dissipation model can be simplified to a linear function that is proportional to the average temperature difference (the difference between the average battery temperature and the ambient temperature).

[0074] In this embodiment, by comprehensively calculating the multi-dimensional characteristics reflecting the temperature rise trend, heat distribution, and internal heat generation, the technical problem of traditional methods being unable to comprehensively and early capture complex thermal runaway precursor signals due to their single characteristics is solved.

[0075] The above are merely feasible implementations of step S20 provided in this embodiment. This embodiment does not specifically limit the specific implementation of step S20.

[0076] Step S30: Based on the thermal runaway characteristics, adjust the dynamic threshold corresponding to the thermal runaway characteristics to obtain the adjusted dynamic threshold;

[0077] It should be noted that the dynamic threshold corresponding to the thermal runaway feature is a critical value used to determine whether the feature triggers a risk alarm; adjusting the dynamic threshold refers to the value obtained after updating the initial dynamic threshold according to the actual operating state of the system.

[0078] For example, dynamic thresholds may include warning thresholds (high-risk thresholds) and critical thresholds.

[0079] It is understandable that, since the operating conditions and environmental conditions of the battery system change over time, fixed thresholds and weights cannot adapt to such dynamic changes. Therefore, step S30 is performed to ensure that the judgment criteria match the current state of the system, avoid the decrease in warning sensitivity or the increase in false alarm rate caused by parameter fixation, and thus improve the adaptability and reliability of the system in long-term operation.

[0080] In one feasible implementation, step S30 may include: obtaining the exponential forgetting coefficient, historical exponential forgetting mean, and historical forgetting variance corresponding to the thermal runaway feature; when the thermal runaway feature does not meet the anomaly removal conditions, obtaining the current exponential forgetting mean based on the historical exponential forgetting mean, the exponential forgetting coefficient, and the thermal runaway feature; obtaining the current forgetting variance based on the thermal runaway feature, the historical exponential forgetting mean, the exponential forgetting coefficient, and the historical forgetting variance; and adjusting the dynamic threshold corresponding to the thermal runaway feature based on the current exponential forgetting mean and the current forgetting variance to obtain the adjusted dynamic threshold.

[0081] It should be noted that the exponential forgetting coefficient is a smoothing parameter used to control the degree of influence of historical data; the historical exponential forgetting mean refers to the average value of the feature calculated by the exponential forgetting method in the previous period; the historical forgetting variance refers to the degree of feature fluctuation calculated by the exponential forgetting method in the previous period; the outlier removal condition refers to the outlier judgment rule set based on the historical variance multiple; the current exponential forgetting mean refers to the average value of the feature obtained after updating with the current feature value; and the current forgetting variance refers to the degree of feature fluctuation obtained after updating with the current feature value.

[0082] Specifically, adjusting the dynamic threshold corresponding to the thermal runaway feature based on the current exponential forgetting mean and the current forgetting variance to obtain the adjusted dynamic threshold includes: obtaining the operating condition bias, the high-risk classification coefficient, and the critical classification coefficient; obtaining the high-risk classification threshold based on the current exponential forgetting mean, the current forgetting variance, the operating condition bias, and the high-risk classification coefficient; obtaining the critical classification threshold based on the current exponential forgetting mean, the current forgetting variance, the operating condition bias, and the critical classification coefficient; and using the high-risk classification threshold and the critical classification threshold as the adjusted dynamic threshold.

[0083] It should be noted that the operating condition bias refers to the threshold correction amount determined based on the current battery operating mode; the high-risk classification coefficient is a multiplier factor used to calculate the high-risk threshold; the critical classification coefficient is a multiplier factor used to calculate the critical threshold; the high-risk classification threshold is the boundary value for determining whether a feature enters a high-risk state; and the critical classification threshold is the boundary value for determining whether a feature enters a critical state.

[0084] For example, the operating condition bias can be obtained by querying a preset table based on the charging and discharging status; the grading coefficient can be set to a fixed value based on engineering experience.

[0085] The formula for calculating the current index forgetting mean is as follows:

[0086]

[0087] The formula for calculating the current exponential forgetting variance is as follows:

[0088]

[0089] The anomaly removal criteria are as follows:

[0090]

[0091] The formulas for calculating the high-risk classification threshold and the critical classification threshold are as follows:

[0092]

[0093] The parameters of the above formula are explained below:

[0094] Thermal runaway characteristics At any moment The current index forgetting mean represents the latest normal baseline; This is the value from the previous period;

[0095] Thermal runaway characteristics The current index forgetting variance measures the volatility, where... ;

[0096] The instantaneous value of the thermal runaway characteristics calculated at time t is used in the real-time scoring;

[0097] Exponential forgetting coefficient (0.001~0.05), balancing "tracking speed / smoothing";

[0098] Outlier removal factor (generally 3~5), the larger the value, the fewer outliers are removed;

[0099] Grading factor (e.g., 2.0 vs. 3.0 or empirical calibration);

[0100] Operating condition bias (an additional value obtained from the operating mode table, typically with an absolute value <0.5σ);

[0101] Thermal runaway characteristics Dynamic thresholds at high-risk and critical levels.

[0102] In this embodiment, by establishing an adaptive baseline and dynamically updating the feature threshold, the technical problem of reduced early warning accuracy caused by the inability of fixed thresholds to track long-term changes in system status is solved.

[0103] The above are merely feasible implementations of step S30 provided in this embodiment. This embodiment does not specifically limit the specific implementation of step S30.

[0104] Step S40: Determine the thermal runaway risk level based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight.

[0105] It should be noted that the thermal runaway risk level is a quantitative classification of the current degree of thermal runaway danger of the battery system, used to guide the adoption of different levels of safety response measures.

[0106] For example, refer to Figure 2 Thermal runaway risk levels can be classified into multiple levels, such as early warning state, high risk state, and critical state.

[0107] Understandably, since thermal runaway is a gradual process, differentiated response strategies are needed based on different levels of risk. Therefore, step S40 enables a precise quantitative assessment of the risk level, improving the sophistication of safety management and system availability.

[0108] In one feasible implementation, step S40 may include: obtaining a feature risk score based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight; obtaining a comprehensive risk score based on the feature risk score; acquiring a reference bias, current-related slope, and bus current; obtaining a comprehensive risk classification threshold based on the reference bias, current-related slope, and bus current; determining classification conditions based on the thermal runaway characteristics, the adjusted dynamic threshold, and the comprehensive risk classification threshold; and determining the thermal runaway risk level based on the comprehensive risk score, the thermal runaway characteristics, and the classification conditions.

[0109] It should be noted that the characteristic risk score is the characteristic risk value calculated by adjusting each thermal runaway characteristic according to its corresponding weight; the comprehensive risk score is the sum of the characteristic risk scores; the baseline bias is the basic offset of the characteristic risk score; the current correlation slope is the proportional coefficient used to adjust the threshold according to the current magnitude; the bus current is the real-time current value flowing through the main circuit of the battery system; the comprehensive risk classification threshold is the scoring boundary value used to divide different risk levels; the classification conditions are the set of rules that must be met to determine the risk level, including the scoring threshold, characteristic threshold, and duration requirements; and the thermal runaway risk level is the final determined current risk state level of the system.

[0110] For example, as shown in the table below:

[0111]

[0112] Parameter descriptions in the table:

[0113] The highest unit or sensor temperature within the current window, reflecting the location of the hottest spot. The high-risk and critical-risk thresholds corresponding to the hotspot temperatures are calculated based on the mean / variance. ;

[0114] First-order heating rate characteristics, in °C / min. , These correspond to "high-risk / critical" rate thresholds, which are dynamically updated according to the operating mode.

[0115] Temperature gradient / dispersion characteristics, which can be measured by... Indicators such as these indicate that To represent temperature channel data; Identify local hotspot thresholds. Identify the thermal diffusion threshold. This represents the number of judgment periods that need to be maintained continuously.

[0116] Self-heating residual under static conditions Description in Net warming trend under the given conditions; The residual trigger threshold, It is the absolute value of the maximum allowable bus current when the bus is "stationary".

[0117] Risk score after feature fusion Requires continuous satisfaction Number of repetitions to avoid shaking; Depend on calculate, , It is the baseline bias of level k. It is the current-dependent slope of level k. Used to construct hierarchical thresholds It is the current amplitude normalized to the rated current.

[0118] In this embodiment, by constructing a multi-dimensional and adaptive hierarchical judgment system, the technical problems of insufficient reliability of single parameter judgment and coarse risk level classification are solved.

[0119] The above are merely feasible implementations of step S40 provided in this embodiment. This embodiment does not specifically limit the specific implementation of step S40.

[0120] Step S50: Execute the corresponding graded response control strategy according to the thermal runaway risk level.

[0121] It should be noted that the graded response control strategy is a set of progressive system control commands that precisely correspond to different levels of thermal runaway risk.

[0122] For example, the strategy may include specific control actions at multiple levels, ranging from increasing the monitoring frequency to ultimately cutting off the main circuit.

[0123] Understandably, since the development of thermal runaway risk is a process, a single, rigid response method cannot take into account both safety and system availability. Therefore, step S50 can implement response measures with matching intensity according to the level of risk, so as to avoid the accident from escalating due to insufficient response or unnecessary system shutdown due to excessive response, thereby significantly improving the continuous operation capability of the system while ensuring safety.

[0124] In one feasible implementation, step S50 may include: when the thermal runaway risk level is a first risk level, suspending the active battery balancing operation and recording an enhanced log; when the thermal runaway risk level is a second risk level, adding a strategy of limiting the battery charging and discharging power and activating or enhancing the heat dissipation system based on the response control strategy of the first risk level; when the thermal runaway risk level is a third risk level, executing a strategy of cutting off the battery main circuit, maintaining the heat dissipation system at maximum operation, and writing key data into non-volatile memory, wherein the first risk level is lower than the second risk level, and the second risk level is lower than the third risk level.

[0125] It should be noted that the first risk level refers to the early warning level, the second risk level refers to the high risk level, and the third risk level refers to the critical safety level.

[0126] For example, increasing the sampling frequency can increase the period from the usual 1 second to 200 milliseconds; power limiting can limit the charging current to 50% of the rated value; key data includes all feature data, status parameters and timestamps within a specific time window before the event is triggered.

[0127] In this embodiment, by establishing a progressive response mechanism that strictly corresponds to the risk level, the technical problem that traditional solutions cannot achieve a balance between security and system availability due to their single response strategy is solved.

[0128] In one feasible implementation, step S50 may subsequently include: calculating the contribution of events with a thermal runaway risk level of the second or third risk level; and updating the adjustment weights based on the contribution to obtain updated adjustment weights.

[0129] It should be noted that an event refers to a complete record of all thermal runaway characteristic data, state parameters, and control commands recorded by the system from the moment the risk level exceeds the preset threshold until the end of that level or the upgrade to a higher level; contribution refers to the cumulative number of second- or third-risk-level events experienced by the system within a preset statistical period.

[0130] For example, the statistical period can be set to several complete charge-discharge cycles of the vehicle, or several hours of continuous system operation; when the frequency of a certain type of risk level event is significantly higher than the historical average, it indicates that the current feature weight settings may need to be optimized.

[0131] In this embodiment, a macro-statistical mechanism based on the frequency of risk events is introduced to assist in adjusting the weights. This solves the technical problem that relying solely on the contribution analysis of a single event may be affected by accidental factors, leading to an insufficiently robust weight update, thereby improving the long-term stability and adaptability of the self-learning process.

[0132] The above are merely feasible implementations of step S50 provided in this embodiment. This embodiment does not specifically limit the specific implementation of step S50.

[0133] This embodiment provides a multi-feature fusion early warning method for battery thermal runaway. The method acquires temperature channel data, ambient temperature, and adjustment weights from the battery system. Based on the temperature channel data and ambient temperature, thermal runaway features are obtained. Based on the thermal runaway features, dynamic thresholds corresponding to the thermal runaway features are adjusted to obtain adjusted dynamic thresholds. Based on the thermal runaway features, the adjusted dynamic thresholds, and the adjustment weights, the thermal runaway risk level is determined. According to the thermal runaway risk level, a corresponding graded response control strategy is executed. By employing a technique that dynamically updates feature thresholds and fusion weights based on real-time operational data, the method solves the technical problems of inaccurate early warning sensitivity and high false alarm rates in existing fixed-threshold methods under battery aging, environmental changes, and dynamic operating conditions. This enables earlier and more accurate identification of thermal runaway risks and supports a progressive safety response from early warning to shutdown, thereby improving safety while ensuring system availability.

[0134] For example, to help understand the implementation process of the battery thermal runaway multi-feature fusion early warning method obtained by combining this embodiment with the above embodiment one, please refer to Figure 3 , Figure 3A simplified flowchart of a multi-feature fusion early warning method for battery thermal runaway is provided, specifically:

[0135] The thermal runaway early warning process in this embodiment begins at the data input layer, and gradually achieves risk identification and response through multiple collaborative modules. First, the system acquires the battery system's temperature array, ambient temperature, and bus current data. Then, the feature extraction module calculates the first-order heating rate, hotspot temperature, temperature gradient characteristics, and self-heating residual during the resting period as thermal runaway features based on the temperature data. Figure 3 The second-order heating acceleration is obtained based on the first-order heating rate; at the same time, the adaptive baseline module updates the mean and variance of each feature based on the exponential forgetting method, and calculates the dynamic threshold accordingly. During the process, outliers exceeding the preset range are removed to ensure that the baseline represents the true state of the system.

[0136] After obtaining the thermal runaway characteristics and dynamic thresholds, the classification judgment module determines the thermal runaway risk level based on the thermal runaway characteristics, the adjusted dynamic thresholds, and the adjusted weights. Specifically, the module compares each characteristic value with its corresponding threshold and combines them with the weights to form a characteristic risk score. Based on preset classification conditions, the risk is divided into multiple levels. The classification conditions include whether the characteristic value reaches its critical threshold and whether the comprehensive score continuously exceeds the threshold, thereby achieving accurate risk state mapping.

[0137] Finally, the response control module executes a graded response control strategy based on the determined thermal runaway risk level. For example, at a low risk level, the sampling frequency is adjusted and equalization is paused; at a medium risk level, power limiting and heat dissipation are increased; and at a high risk level, the main loop is cut off and data is solidified. Throughout the process, the self-learning module dynamically updates the weights based on the characteristic contribution of historical warning events, and the event log module records key data for retrospection and analysis, thus forming a complete closed-loop processing flow from data perception, feature analysis, dynamic evaluation to graded execution.

[0138] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the multi-feature fusion early warning method for battery thermal runaway in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0139] This application also provides a battery thermal runaway multi-feature fusion early warning device, please refer to... Figure 4 The battery thermal runaway multi-feature fusion early warning device includes:

[0140] The acquisition module 10 is used to acquire temperature channel data, ambient temperature, and adjustment weights in the battery system.

[0141] Feature module 20 is used to obtain thermal runaway features based on the temperature channel data and the ambient temperature;

[0142] The adjustment module 30 is used to adjust the dynamic threshold corresponding to the thermal runaway feature based on the thermal runaway feature, so as to obtain the adjusted dynamic threshold.

[0143] The grading module 40 is used to determine the thermal runaway risk level based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight.

[0144] The execution module 50 is used to execute a corresponding graded response control strategy based on the thermal runaway risk level.

[0145] The battery thermal runaway multi-feature fusion early warning device provided in this application, employing the battery thermal runaway multi-feature fusion early warning method in the above embodiments, can solve the technical problem of how to improve the timeliness and accuracy of battery thermal runaway early warning under dynamic operating conditions. Compared with the prior art, the beneficial effects of the battery thermal runaway multi-feature fusion early warning device provided in this application are the same as those of the battery thermal runaway multi-feature fusion early warning method provided in the above embodiments, and other technical features in the battery thermal runaway multi-feature fusion early warning device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0146] The feature module 20 is further configured to obtain, based on the temperature channel data, hotspot temperature, temperature gradient characteristics, window average temperature, periodic average temperature, and multiple first-order heating rates; acquire the quiescent heating rate and heat dissipation model; obtain the quiescent section self-heating residual based on the quiescent heating rate, the heat dissipation model, the periodic average temperature, and the ambient temperature; and use the window average temperature, the first-order heating rate, the hotspot temperature, the temperature gradient characteristics, and the quiescent section self-heating residual as thermal runaway characteristics.

[0147] The adjustment module 30 is further configured to obtain the exponential forgetting coefficient, historical exponential forgetting mean, and historical forgetting variance corresponding to the thermal runaway feature; when the thermal runaway feature does not meet the anomaly removal conditions, obtain the current exponential forgetting mean based on the historical exponential forgetting mean, the exponential forgetting coefficient, and the thermal runaway feature; obtain the current forgetting variance based on the thermal runaway feature, the historical exponential forgetting mean, the exponential forgetting coefficient, and the historical forgetting variance; and adjust the dynamic threshold corresponding to the thermal runaway feature based on the current exponential forgetting mean and the current forgetting variance to obtain the adjusted dynamic threshold.

[0148] The adjustment module 30 is further configured to obtain the operating condition bias, high-risk classification coefficient, and critical classification coefficient; obtain a high-risk classification threshold based on the current index forgetting mean, the current forgetting variance, the operating condition bias, and the high-risk classification coefficient; obtain a critical classification threshold based on the current index forgetting mean, the current forgetting variance, the operating condition bias, and the critical classification coefficient; and use the high-risk classification threshold and the critical classification threshold as adjustment dynamic thresholds.

[0149] The grading module 40 is further configured to: obtain a feature risk score based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight; obtain a comprehensive risk score based on the feature risk score; acquire a reference bias, current-related slope, and bus current; obtain a comprehensive risk grading threshold based on the reference bias, current-related slope, and bus current; determine grading conditions based on the thermal runaway characteristics, the adjusted dynamic threshold, and the comprehensive risk grading threshold; and determine the thermal runaway risk level based on the comprehensive risk score, the thermal runaway characteristics, and the grading conditions.

[0150] The execution module 50 is further configured to: suspend the active battery balancing operation and record an enhanced log when the thermal runaway risk level is the first risk level; add a strategy of limiting the battery charging and discharging power and activating or enhancing the heat dissipation system based on the response control strategy of the first risk level when the thermal runaway risk level is the second risk level; and execute a strategy of cutting off the battery main circuit, maximizing the operation of the heat dissipation system, and writing key data into a non-volatile memory when the thermal runaway risk level is the third risk level, wherein the first risk level is lower than the second risk level, and the second risk level is lower than the third risk level.

[0151] The execution module 50 is further configured to calculate the contribution of events with a thermal runaway risk level of the second or third risk level; and update the adjustment weights based on the contribution to obtain the updated adjustment weights.

[0152] This application provides a battery thermal runaway multi-feature fusion early warning device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the battery thermal runaway multi-feature fusion early warning method in the above embodiment 1.

[0153] The following is for reference. Figure 5This document illustrates a structural schematic diagram suitable for implementing a battery thermal runaway multi-feature fusion early warning device according to embodiments of this application. The battery thermal runaway multi-feature fusion early warning device in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 5 The battery thermal runaway multi-feature fusion early warning device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0154] like Figure 5 As shown, the battery thermal runaway multi-feature fusion early warning device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into random access memory (RRAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the battery thermal runaway multi-feature fusion early warning device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the battery thermal runaway multi-feature fusion early warning device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a battery thermal runaway multi-feature fusion early warning device with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0155] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0156] The battery thermal runaway multi-feature fusion early warning device provided in this application, employing the battery thermal runaway multi-feature fusion early warning method in the above embodiments, can solve the technical problem of how to improve the timeliness and accuracy of battery thermal runaway early warning under dynamic operating conditions. Compared with the prior art, the beneficial effects of the battery thermal runaway multi-feature fusion early warning device provided in this application are the same as the beneficial effects of the battery thermal runaway multi-feature fusion early warning method provided in the above embodiments, and other technical features in this battery thermal runaway multi-feature fusion early warning device are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0157] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0158] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0159] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the battery thermal runaway multi-feature fusion early warning method in the above embodiments.

[0160] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0161] The aforementioned computer-readable storage medium may be included in the battery thermal runaway multi-feature fusion early warning device; or it may exist independently and not assembled into the battery thermal runaway multi-feature fusion early warning device.

[0162] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by the battery thermal runaway multi-feature fusion early warning device, the battery thermal runaway multi-feature fusion early warning device: acquires temperature channel data, ambient temperature, and adjustment weights in the battery system; obtains thermal runaway features based on the temperature channel data and the ambient temperature; adjusts the dynamic threshold corresponding to the thermal runaway features based on the thermal runaway features to obtain an adjusted dynamic threshold; determines the thermal runaway risk level based on the thermal runaway features, the adjusted dynamic threshold, and the adjustment weights; and executes a corresponding graded response control strategy according to the thermal runaway risk level.

[0163] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0164] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0165] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0166] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described battery thermal runaway multi-feature fusion early warning method. This solves the technical problem of how to improve the timeliness and accuracy of battery thermal runaway early warning under dynamic operating conditions. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the battery thermal runaway multi-feature fusion early warning method provided in the above embodiments, and will not be repeated here.

[0167] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the battery thermal runaway multi-feature fusion early warning method described above.

[0168] The computer program product provided in this application can solve the technical problem of how to improve the timeliness and accuracy of battery thermal runaway early warning under dynamic operating conditions. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the battery thermal runaway multi-feature fusion early warning method provided in the above embodiments, and will not be repeated here.

[0169] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A multi-feature fusion early warning method for battery thermal runaway, characterized in that, The method includes: Acquire temperature channel data, ambient temperature, and adjustment weights in the battery system; Based on the temperature channel data and the ambient temperature, the thermal runaway characteristics are obtained; Based on the thermal runaway characteristics, the dynamic threshold corresponding to the thermal runaway characteristics is adjusted to obtain the adjusted dynamic threshold; Based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight, the thermal runaway risk level is determined. Based on the aforementioned thermal runaway risk level, implement the corresponding graded response control strategy; The step of adjusting the dynamic threshold corresponding to the thermal runaway characteristics based on the thermal runaway characteristics to obtain the adjusted dynamic threshold includes: Obtain the exponential forgetting coefficient, historical exponential forgetting mean, and historical forgetting variance corresponding to the thermal runaway characteristics; When the thermal runaway features do not meet the anomaly removal criteria, the current exponential forgetting mean is obtained based on the historical exponential forgetting mean, the exponential forgetting coefficient, and the thermal runaway features. Based on the thermal runaway characteristics, the historical exponential forgetting mean, the exponential forgetting coefficient, and the historical forgetting variance, the current forgetting variance is obtained; Based on the current exponential forgetting mean and the current forgetting variance, the dynamic threshold corresponding to the thermal runaway feature is adjusted to obtain the adjusted dynamic threshold.

2. The method as described in claim 1, characterized in that, The process of obtaining thermal runaway characteristics based on the temperature channel data and the ambient temperature includes: Based on the temperature channel data, hotspot temperature, temperature gradient characteristics, window average temperature, period average temperature, and multiple first-order heating rates are obtained. The temperature gradient characteristics are an indicator used to quantify the uniformity of temperature distribution inside the battery system. The window average temperature is the arithmetic mean of the instantaneous temperature values ​​of all effective temperature channels within a preset sliding time window. The period average temperature is the average of the instantaneous readings of all temperature channels within a single sampling period. The first-order heating rate is the amount of temperature change per unit time. Obtain the heating rate and heat dissipation model under static operating conditions; Based on the heating rate under static conditions, the heat dissipation model, the periodic average temperature, and the ambient temperature, the self-heating residual of the static section is obtained. The average window temperature, the first-order heating rate, the hot spot temperature, the temperature gradient characteristics, and the self-heating residual of the static section are used as thermal runaway characteristics.

3. The method as described in claim 1, characterized in that, The step of adjusting the dynamic threshold corresponding to the thermal runaway feature based on the current exponential forgetting mean and the current forgetting variance, to obtain the adjusted dynamic threshold, includes: Obtain operating condition bias, high-risk classification coefficient, and critical classification coefficient, wherein the operating condition bias refers to the threshold correction amount determined based on the current battery operating mode; The high-risk classification threshold is obtained based on the current index forgetting mean, the current forgetting variance, the operating condition bias, and the high-risk classification coefficient. The critical grading threshold is obtained based on the current index forgetting mean, the current forgetting variance, the operating condition bias, and the critical grading coefficient. The high-risk classification threshold and the critical classification threshold are used as the dynamic adjustment thresholds.

4. The method as described in claim 1, characterized in that, The determination of the thermal runaway risk level based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight includes: Based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight, a characteristic risk score is obtained; Based on the aforementioned feature risk scores, a comprehensive risk score is obtained; Obtain the reference bias, current correlation slope, and bus current, wherein the reference bias is the basic offset for feature risk scoring, and the current correlation slope is a proportional coefficient used to adjust the threshold according to the current magnitude. Based on the aforementioned reference bias, current-related slope, and bus current, a comprehensive risk classification threshold is obtained; Based on the thermal runaway characteristics, the adjusted dynamic threshold, and the comprehensive risk classification threshold, the classification conditions are determined. Based on the comprehensive risk score, the thermal runaway characteristics, and the grading conditions, the thermal runaway risk level is determined.

5. The method as described in claim 1, characterized in that, The step of implementing a corresponding graded response control strategy based on the thermal runaway risk level includes: When the thermal runaway risk level is the first risk level, the strategy is to suspend the active battery balancing operation and record the enhanced log. When the thermal runaway risk level is the second risk level, based on the response control strategy of the first risk level, a strategy of limiting the battery charging and discharging power and activating or strengthening the heat dissipation system is added. When the thermal runaway risk level is the third risk level, a strategy is implemented to disconnect the battery main circuit, maintain the heat dissipation system at maximum operation, and write critical data into non-volatile memory. The first risk level is lower than the second risk level, and the second risk level is lower than the third risk level.

6. The method as described in claim 1, characterized in that, After executing the corresponding graded response control strategy based on the thermal runaway risk level, the process includes: The contribution of events with a thermal runaway risk level of the second or third risk level is statistically analyzed. The adjustment weights are updated based on the contribution level to obtain the updated adjustment weights.

7. A battery thermal runaway multi-feature fusion early warning device, characterized in that, The device includes: The acquisition module is used to acquire temperature channel data, ambient temperature, and adjustment weights in the battery system. The feature module is used to obtain thermal runaway features based on the temperature channel data and the ambient temperature; An adjustment module is used to adjust the dynamic threshold corresponding to the thermal runaway feature based on the thermal runaway feature, so as to obtain the adjusted dynamic threshold. The grading module is used to determine the thermal runaway risk level based on the thermal runaway characteristics, the adjusted dynamic threshold, and the adjusted weight. The execution module is used to execute a corresponding graded response control strategy based on the thermal runaway risk level. The adjustment module is further configured to obtain the exponential forgetting coefficient, historical exponential forgetting mean, and historical forgetting variance corresponding to the thermal runaway feature; when the thermal runaway feature does not meet the anomaly removal conditions, obtain the current exponential forgetting mean based on the historical exponential forgetting mean, the exponential forgetting coefficient, and the thermal runaway feature; obtain the current forgetting variance based on the thermal runaway feature, the historical exponential forgetting mean, the exponential forgetting coefficient, and the historical forgetting variance; and adjust the dynamic threshold corresponding to the thermal runaway feature based on the current exponential forgetting mean and the current forgetting variance to obtain the adjusted dynamic threshold.

8. A battery thermal runaway multi-feature fusion early warning device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the battery thermal runaway multi-feature fusion early warning method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the battery thermal runaway multi-feature fusion early warning method as described in any one of claims 1 to 6.

Images