Battery pack thermal runaway probability determination method, device and electronic equipment

Abstract

The application discloses a battery pack thermal runaway probability determination method and device and electronic equipment. The method comprises the following steps: collecting multiple sets of operation data of the battery pack in a predetermined time period, wherein the multiple sets of operation data correspond to multiple sub-time periods included in the predetermined time period in a one-to-one manner; for any sub-time period in the multiple sub-time periods, determining a target thermal runaway flag of the battery pack in the any sub-time period based on the operation data of the any sub-time period, wherein the target thermal runaway flag is used to indicate whether the battery pack has a thermal runaway risk in the any sub-time period; and determining a thermal runaway probability of the battery pack in the predetermined time period based on the target thermal runaway flags corresponding to the multiple sub-time periods, wherein the target thermal runaway flags corresponding to the multiple sub-time periods are obtained by using the manner of determining the target thermal runaway flag of the any sub-time period. The application solves the technical problem of low accuracy of the thermal runaway probability determination result of the battery pack in the related art.

Description

Technical Field

[0001] This application relates to the field of battery technology, and more specifically, to a method, apparatus, and electronic device for determining the thermal runaway probability of a battery pack. Background Technology

[0002] As the number of electric vehicles (EVs) increases, their safety is receiving growing attention. Due to the highly reactive chemical properties of lithium-ion battery materials, EVs are prone to thermal runaway. Therefore, determining the probability of thermal runaway in the battery pack and subsequently providing early warning systems for EVs is crucial for their safety. However, thermal runaway in battery packs can be caused by a variety of factors, such as overcharging / over-discharging, internal short circuits, external high temperatures, and electrolyte decomposition. It can also be caused by a combination of factors, making early warning systems for EV thermal runaway challenging.

[0003] In related technologies, machine learning models are typically used to determine the probability of thermal runaway in battery packs. However, this method is overly reliant on the quality of the collected data, and the model's generalization ability is limited. Furthermore, due to the complexity and diversity of the causes of thermal runaway in battery packs, the aforementioned methods tend to have significant errors in determining the probability of thermal runaway. Therefore, related technologies suffer from low accuracy in determining the probability of thermal runaway in battery packs.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a method, apparatus, and electronic device for determining the thermal runaway probability of a battery pack, in order to at least solve the technical problem of low accuracy in determining the thermal runaway probability of a battery pack in related technologies.

[0006] According to one aspect of the embodiments of this application, a method for determining the thermal runaway probability of a battery pack is provided, comprising: collecting multiple sets of operating data of the battery pack over a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period; for any sub-time period, determining a target thermal runaway indicator of the battery pack in that sub-time period based on the operating data of that sub-time period, wherein the target thermal runaway indicator is used to indicate whether there is a risk of thermal runaway in the battery pack in that sub-time period; and determining the thermal runaway probability of the battery pack in the predetermined time period based on the target thermal runaway indicators corresponding to the multiple sub-time periods respectively, wherein the target thermal runaway indicators corresponding to the multiple sub-time periods are obtained by determining the target thermal runaway indicator of any sub-time period. Calculating the thermal runaway probability of the battery pack in the predetermined time period by using the target thermal runaway indicators corresponding to the battery pack in the multiple sub-time periods can provide a quantitative assessment of the thermal runaway risk of the battery pack, improve the accuracy of the determination result of the thermal runaway probability of the battery pack, and thus realize early warning and prevention measures for thermal runaway of the battery pack.

[0007] Optionally, based on the operational data of any sub-time period, the target thermal runaway indicator of the battery pack in any sub-time period is determined, including: determining high-dimensional datasets corresponding to multiple operational states of the battery pack in any sub-time period based on the operational data; determining the target centroid information corresponding to multiple high-dimensional datasets in any sub-time period, wherein multiple high-dimensional datasets correspond one-to-one with multiple operational states; and determining the target thermal runaway indicator of the battery pack in any sub-time period based on the target centroid information corresponding to multiple high-dimensional datasets. By determining the centroid information of the high-dimensional datasets, key information of the battery pack in the corresponding operational states can be determined, laying the foundation for the subsequent determination of the probability of thermal runaway of the battery pack.

[0008] Optionally, based on operational data, high-dimensional datasets corresponding to multiple operational states of the battery pack in any sub-time period are determined, including: preprocessing the operational data to obtain standard data of the battery pack in any sub-time period, wherein the preprocessing includes at least one of the following: data cleaning, data filling, and data splitting; segmenting the standard data according to operational state judgment logic to obtain standard data corresponding to multiple operational states, wherein the operational state judgment logic is used to identify and classify data of the battery pack in different operational states; determining statistical features corresponding to multiple operational states based on the standard data corresponding to multiple operational states; and determining high-dimensional datasets corresponding to multiple operational states based on the standard data corresponding to multiple operational states and the statistical features corresponding to multiple operational states. Preprocessing the operational data can improve the quality of the operational data and reduce the impact of data noise on the determination of the thermal runaway probability. Simultaneously, segmenting the operational data according to operational states allows for more accurate identification of the characteristics of the battery pack in different operational states, providing a rich state perspective for determining the thermal runaway probability of the battery pack.

[0009] Optionally, determining the target centroid information corresponding to multiple high-dimensional datasets for any sub-time period includes: for any operating state among multiple operating states, clustering the high-dimensional dataset of any state for any operating state to obtain the centroid of any state in the high-dimensional dataset of any state; extracting information from the centroid of any state to obtain multiple initial centroid information for the centroid of any state; fusing the multiple initial centroid information to obtain the target centroid information of any state in the high-dimensional dataset of any state; and obtaining the target centroid information corresponding to multiple operating states by using the method of determining the target centroid information of any state. By fusing the information of multiple initial centroids, the uncertainty caused by the random fluctuation of a single initial centroid can be reduced, and the accuracy of the determination result of the probability of thermal runaway of the battery pack can be improved.

[0010] Optionally, based on the target centroid information corresponding to multiple high-dimensional datasets, the target thermal runaway indicator of the battery pack in any sub-time period is determined. This includes: determining the average centroid spacing and the first-order centroid difference for each of the multiple high-dimensional datasets, where the average centroid spacing represents the average distance between the data points in the high-dimensional datasets and the centroids, and the first-order centroid difference represents the change of the centroids over time. Based on the average centroid spacing and the first-order centroid difference for each of the multiple high-dimensional datasets, the target thermal runaway indicator of the battery pack in any sub-time period is determined. Determining the average centroid spacing and the first-order centroid difference not only identifies the deviation between the data points in the high-dimensional datasets and the centroids of the high-dimensional datasets, but also determines the trend of the battery pack's state change over time, improving the accuracy of the battery pack thermal runaway probability determination results.

[0011] Optionally, based on the target centroid information corresponding to multiple high-dimensional datasets, the average centroid spacing corresponding to each of the multiple high-dimensional datasets is determined, including: determining the first centroid spacing corresponding to each of the multiple high-dimensional datasets in any sub-time period based on the target centroid information corresponding to each of the multiple high-dimensional datasets; determining the second centroid spacing corresponding to each of the multiple high-dimensional datasets in the historical sub-time period based on the historical target centroid information of the historical sub-time period; and determining the average centroid spacing corresponding to each of the multiple high-dimensional datasets based on the first centroid spacing corresponding to each of the multiple high-dimensional datasets in any sub-time period and the second centroid spacing corresponding to each of the multiple high-dimensional datasets in the historical sub-time period. Combining the second centroid spacing of the historical sub-time period with the first centroid spacing of any sub-time period allows for the identification of the trend in the state of the battery pack over time, thereby identifying early signs of thermal runaway of the battery pack and helping to improve the accuracy of the determination of the probability of thermal runaway of the battery pack.

[0012] Optionally, based on the average centroid spacing and the first-order difference of the centroids corresponding to multiple high-dimensional datasets, the target thermal runaway indicator of the battery pack in any sub-time period is determined. This includes: obtaining a first thermal runaway indicator based on the average centroid spacing and a pre-set first threshold; obtaining a second thermal runaway indicator based on the first-order difference of the centroids and a pre-set second threshold; and determining the target thermal runaway indicator based on the first and second thermal runaway indicators. By combining the average centroid spacing and the first-order difference of the centroids to determine the target thermal runaway indicator of the battery pack, compared to judging with a single indicator, the thermal runaway state of the battery pack can be assessed more comprehensively, improving the accuracy of the determination of the thermal runaway state of the battery pack, and thus improving the accuracy of the determination of the thermal runaway probability of the battery pack.

[0013] Optionally, based on the target thermal runaway markers corresponding to multiple sub-time periods, the probability of thermal runaway of the battery pack within a predetermined time period is determined. This includes: determining a first number of target thermal runaway markers indicating a risk of thermal runaway among the target thermal runaway markers corresponding to the multiple sub-time periods, and a second number of sub-time periods included in the predetermined time period; and determining the thermal runaway probability based on the first number and the second number. By statistically analyzing the number of target thermal runaway markers identified as having a risk of thermal runaway within the predetermined time period and calculating the probability of thermal runaway of the battery pack within the predetermined time period, a quantitative analysis of the thermal runaway risk of the battery pack can be achieved, improving the accuracy of the determination result of the battery pack's thermal runaway probability.

[0014] According to another aspect of the embodiments of this application, a device for determining the thermal runaway probability of a battery pack is provided, comprising: a data acquisition module, configured to acquire multiple sets of operating data of the battery pack during a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period; a target thermal runaway flag determination module, configured to determine a target thermal runaway flag of the battery pack in any sub-time period based on the operating data of any sub-time period, wherein the target thermal runaway flag is used to indicate whether there is a risk of thermal runaway in the battery pack during any sub-time period; and a thermal runaway probability determination module, configured to determine the thermal runaway probability of the battery pack during the predetermined time period based on the target thermal runaway flags corresponding to the multiple sub-time periods respectively, wherein the target thermal runaway flags corresponding to the multiple sub-time periods are obtained by determining the target thermal runaway flag of any sub-time period.

[0015] According to another aspect of the embodiments of this application, a non-volatile storage medium is provided, which stores a plurality of instructions adapted for a method for determining the thermal runaway probability of a battery pack, any one of which is loaded by a processor.

[0016] According to another aspect of the embodiments of this application, an electronic device is provided, including: one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement any one of the following methods for determining the thermal runaway probability of a battery pack.

[0017] According to another aspect of the embodiments of this application, a computer program product is provided, which, when executed on a data processing device, is adapted to perform the steps of a method for determining the thermal runaway probability of a battery pack.

[0018] In this embodiment, multiple sets of operational data of the battery pack are collected over a predetermined time period, each set corresponding to a sub-time period. For any sub-time period, a target thermal runaway indicator is determined based on the operational data of that sub-time period. This target thermal runaway indicator indicates whether the battery pack has a risk of thermal runaway in any sub-time period. Based on the target thermal runaway indicators corresponding to the sub-time periods, the probability of thermal runaway of the battery pack within the predetermined time period is determined. The target thermal runaway indicators corresponding to the sub-time periods are obtained by determining the target thermal runaway indicator for each sub-time period. This achieves the goal of improving the accuracy of the determination of the battery pack's thermal runaway probability by collecting and analyzing the operational data of the battery pack over a predetermined time period, determining the target thermal runaway risk indicators corresponding to the sub-time periods within the predetermined time period, and thus obtaining the probability of thermal runaway of the battery pack within the predetermined time period. This solves the technical problem of low accuracy in determining the probability of thermal runaway of the battery pack in related technologies. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a flowchart of a method for determining the thermal runaway probability of a battery pack according to an embodiment of this application;

[0021] Figure 2 This is a first flowchart of an optional method for determining the thermal runaway probability of a battery pack according to an embodiment of this application;

[0022] Figure 3 This is a second flowchart of an optional method for determining the thermal runaway probability of a battery pack according to an embodiment of this application;

[0023] Figure 4 This is a schematic diagram of an optional battery pack thermal runaway probability determination device provided according to an embodiment of this application. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] According to an embodiment of this application, a method embodiment for determining the thermal runaway probability of a battery pack is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0027] Figure 1 This is a flowchart of a method for determining the thermal runaway probability of a battery pack according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0028] Step S102: Collect multiple sets of operating data of the battery pack during a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period;

[0029] It is understood that data is collected from the battery pack within a predetermined time period to obtain multiple sets of operational data. This predetermined time period includes multiple sub-time periods, with each set of operational data corresponding to one of these sub-time periods. Through precise and comprehensive data collection, operational data on various operating states of the battery pack within the predetermined time period is obtained, avoiding errors in determining the probability of thermal runaway of the battery pack due to data gaps.

[0030] Step S104: For any sub-time period among multiple sub-time periods, based on the operating data of any sub-time period, determine the target thermal runaway flag of the battery pack in any sub-time period, wherein the target thermal runaway flag is used to indicate whether there is a risk of thermal runaway in the battery pack in any sub-time period.

[0031] It is understandable that, for any given sub-time period among multiple sub-time periods, the operational data of the battery pack in that sub-time period is analyzed and processed to determine the target thermal runaway indicator of the battery pack in that sub-time period. By analyzing and processing the operational data of the battery pack in any sub-time period, the target thermal runaway indicator of the battery pack in that sub-time period is obtained, thus laying the foundation for determining the probability of thermal runaway of the battery pack in a predetermined time period.

[0032] In one optional embodiment, determining the target thermal runaway indicator of the battery pack in any sub-time period based on the operational data of any sub-time period includes: determining high-dimensional datasets corresponding to multiple operational states of the battery pack in any sub-time period based on the operational data; determining target centroid information corresponding to multiple high-dimensional datasets in any sub-time period, wherein the multiple high-dimensional datasets correspond one-to-one with multiple operational states; and determining the target thermal runaway indicator of the battery pack in any sub-time period based on the target centroid information corresponding to the multiple high-dimensional datasets.

[0033] It is understandable that analyzing the operational data of the battery pack in any sub-time period yields high-dimensional datasets corresponding to multiple operational states of the battery pack in that sub-time period. Clustering is then performed on these high-dimensional datasets to obtain target centroid information for each dataset. Based on this target centroid information, target thermal runaway markers are determined to indicate whether the battery pack faces a risk of thermal runaway in any sub-time period. By determining the centroid information of the high-dimensional datasets, key information about the battery pack in the corresponding operational state can be identified, laying the foundation for subsequently determining the probability of thermal runaway of the battery pack.

[0034] In one optional embodiment, determining the high-dimensional datasets corresponding to multiple operating states of the battery pack in any sub-time period based on operational data includes: preprocessing the operational data to obtain standard data of the battery pack in any sub-time period, wherein the preprocessing includes at least one of the following: data cleaning, data filling, and data splitting; segmenting the standard data according to the operating state judgment logic to obtain standard data corresponding to multiple operating states, wherein the operating state judgment logic is used to identify and classify the data of the battery pack in different operating states; determining the statistical features corresponding to the multiple operating states based on the standard data corresponding to the multiple operating states; and determining the high-dimensional datasets corresponding to the multiple operating states based on the standard data corresponding to the multiple operating states and the statistical features corresponding to the multiple operating states.

[0035] It is understandable that preprocessing the battery pack's operational data for any sub-time period—such as data cleaning, data filling, and data splitting—results in standard data for that sub-time period. This standard data is then segmented according to the operational state judgment logic, yielding standard data corresponding to multiple operational states of the battery pack for each sub-time period. Statistical analysis is then performed on the standard data corresponding to each of these operational states to calculate their respective statistical characteristics, such as mean and variance. Based on the standard data and statistical characteristics corresponding to each operational state, high-dimensional datasets for each operational state are constructed. Preprocessing the operational data improves its quality and reduces the impact of data noise on the determination of thermal runaway probability. Furthermore, segmenting the operational data according to operational states allows for more accurate identification of the battery pack's characteristics under different operational states, providing a richer state perspective for determining the battery pack's thermal runaway probability.

[0036] Optionally, battery pack operating data can be collected from the electric vehicle's Battery Management System (BMS). This data, including voltage, current, insulation resistance, and temperature, is uploaded in real-time to a big data cloud platform. After data preprocessing, time-series data for physical quantities such as current and insulation resistance are obtained. The overall battery pack operating data is then segmented to obtain time-series data for voltage and temperature. These time-series data for current, insulation resistance, voltage, and temperature constitute the preprocessed standard data for the battery pack.

[0037] Optionally, operational data such as voltage, current, insulation resistance, and temperature of the electric vehicle battery pack can be obtained from a big data cloud platform, with a data sampling interval of 30 seconds. This operational data can be preprocessed, including removing outlier data, filling in a small number of missing data, and splitting the voltage and temperature of all cells in the battery pack, to obtain continuous time-series data of the individual cell voltage, temperature, current, and insulation resistance of the battery pack.

[0038] Optionally, voltage can be used as an example to illustrate the method for determining continuous time series data of voltage. Voltage data with intervals exceeding 300 seconds are divided into two different time series datasets. In either time series dataset, for a small number of voltage data points that are 0, missing values ​​are filled using interpolation, and a small number of noisy data points with voltages exceeding a specific range are removed. Finally, the remaining voltage data are organized into a time series to form continuous time series data of voltage.

[0039] Optionally, the standard data of charging state, discharging state and post-charging resting state in the standard data of the preprocessed battery pack obtained above can be identified and tagged to obtain the tagged operating data of the battery pack in a time segment (i.e., any sub-time period), such as 1 day (i.e., standard data corresponding to multiple operating states respectively).

[0040] Optionally, the preprocessed standard data of the battery pack is segmented according to three state judgment logics (i.e., operating state judgment logic) to extract the standard data of the charging state, discharging state, and post-charging static state of the battery pack in the time segment. The standard data of the charging state, discharging state, and post-charging static state are then labeled (Charge, corresponding to the charging state; Discharge, corresponding to the discharging state; Charge_still, corresponding to the post-charging static state).

[0041] Optionally, the above-mentioned operating status judgment logic includes charging status judgment logic, discharging status judgment logic, and post-charging resting status judgment logic. The charging status judgment logic states that when the electric vehicle is in a resting state, the current is greater than 0, the charging / discharging flag is 1, and the State of Health (SOH) gradually increases. The discharging status judgment logic states that the electric vehicle's speed is constantly changing, the SOH gradually decreases, the current is generally less than 0, and the charging / discharging flag is 0. The post-charging resting status judgment logic states that the electric vehicle's speed is 0, the SOH begins to gradually increase, subsequently reaches a stable state, and the current is greater than 0.

[0042] Optionally, features can be extracted from the labeled operational data of the battery pack obtained above for each time segment, and statistical characteristics such as mean and variance of the labeled operational data of the battery pack for each time segment can be calculated. The statistical characteristics of the labeled operational data of the battery pack for each time segment, together with the continuous time series data of the battery pack, form a high-dimensional dataset of the battery pack for each time segment. The size of the high-dimensional dataset can be dynamically changed based on the size of the battery pack.

[0043] In one optional embodiment, determining the target centroid information corresponding to multiple high-dimensional datasets for any sub-time period includes: for any running state among multiple running states, performing clustering processing on the high-dimensional dataset of any state for any running state to obtain the centroid of any state in the high-dimensional dataset of any state; extracting information from the centroid of any state to obtain multiple initial centroid information of the centroid of any state; fusing the multiple initial centroid information to obtain the target centroid information of any state in the high-dimensional dataset of any state; and obtaining the target centroid information corresponding to multiple running states by using the method of determining the target centroid information of any state.

[0044] It can be understood that cluster analysis is performed on any high-dimensional dataset corresponding to any operating state in multiple operating states to obtain the centroid of any state in the high-dimensional dataset of any state, where the dimension of the centroid is the same as the dimension of the high-dimensional dataset of any state. Various information extraction methods (such as t-distributed random neighborhood embedding and singular value decomposition) are used to extract information from the centroid of any state, resulting in multiple initial centroid information. These multiple initial centroid information are then fused to obtain the target centroid information of any state in the high-dimensional dataset of any state. By determining the target centroid information of any state, the target centroid information corresponding to multiple high-dimensional datasets of the battery pack in any sub-time period is obtained. By fusing the information from multiple initial centroids, the uncertainty caused by the random fluctuations of individual initial centroid information can be reduced, improving the accuracy of the determination of the probability of thermal runaway of the battery pack.

[0045] Optionally, clustering can be performed on the high-dimensional datasets corresponding to multiple operating states of the battery pack during time segments. This is because thermal runaway of the battery pack generally has two states: occurring and not occurring, with completely different operational data structures. Furthermore, before thermal runaway occurs in the electric vehicle, the operational data structure of the battery pack will slightly shift towards the operational data structure of thermal runaway. Based on the above analysis, the cluster N of the high-dimensional dataset is set to 2.

[0046] Optionally, based on the above clustering results, centroids corresponding to multiple high-dimensional datasets that correspond one-to-one with multiple operating states are obtained, and these centroids are high-dimensional centroids. Different methods can be used to extract information from the high-dimensional centroids, including but not limited to the t-distribution random neighborhood embedding method and the singular value decomposition method, to extract multiple 2D information of the centroid (i.e., multiple initial centroid information), and after fusion, the main information of the centroid (i.e., target centroid information) is obtained.

[0047] Optionally, multiple initial centroid information is obtained based on the information extraction results. These initial centroid information is then normalized and fused to obtain two-dimensional target centroid information. This two-dimensional target centroid information is mapped onto a two-dimensional space according to the order of sub-time periods, forming two coordinate points in the two-dimensional space (one corresponding to the centroid of the thermal runaway cluster, and the other corresponding to the centroid of the cluster that has not experienced thermal runaway; the distance between the two coordinate points is the centroid spacing of the battery pack in that time segment). The above method is used to obtain two coordinate points of the battery pack in two-dimensional space at different time segments. By analyzing these two coordinate points of the battery pack in two-dimensional space at different time segments, thermal runaway fault warnings can be provided for the battery pack, and the probability of thermal runaway of the battery pack within a predetermined time period can be determined.

[0048] In one optional embodiment, determining the target thermal runaway indicator of the battery pack in any sub-time period based on the target centroid information corresponding to multiple high-dimensional datasets includes: determining the average centroid spacing and the first-order centroid difference corresponding to the multiple high-dimensional datasets based on the target centroid information corresponding to the multiple high-dimensional datasets, wherein the average centroid spacing represents the average distance between the data in the high-dimensional dataset and the centroid, and the first-order centroid difference is used to represent the change of the centroid over time; and determining the target thermal runaway indicator of the battery pack in any sub-time period based on the average centroid spacing and the first-order centroid difference corresponding to the multiple high-dimensional datasets.

[0049] It can be understood that, based on the target centroid information corresponding to multiple high-dimensional datasets, the average centroid spacing and the first-order centroid difference for each high-dimensional dataset are calculated. The average centroid spacing represents the average distance between the data points in the high-dimensional dataset and the centroid, while the first-order centroid difference is the difference between the centroid spacing in any sub-time period and the centroid spacing in adjacent time periods, used to describe the trend of centroid change over time. Based on the average centroid spacing and the first-order centroid difference for each high-dimensional dataset, the target thermal runaway indicator of the battery pack in any sub-time period is determined. The determination of the average centroid spacing and the first-order centroid difference not only identifies the deviation between the data points in the high-dimensional dataset and the centroid of the high-dimensional dataset, but also determines the trend of battery pack state change over time, improving the accuracy of the battery pack thermal runaway probability determination results.

[0050] In one optional embodiment, determining the average centroid spacing of the multiple high-dimensional datasets based on the target centroid information corresponding to the multiple high-dimensional datasets respectively includes: determining a first centroid spacing of the multiple high-dimensional datasets corresponding to any sub-time period based on the target centroid information corresponding to the multiple high-dimensional datasets respectively; determining a second centroid spacing of the battery pack corresponding to the multiple high-dimensional datasets in the historical sub-time period based on the historical target centroid information of the historical sub-time period before any sub-time period; and determining the average centroid spacing of the multiple high-dimensional datasets based on the first centroid spacing of the multiple high-dimensional datasets corresponding to any sub-time period and the second centroid spacing of the multiple high-dimensional datasets corresponding to the multiple high-dimensional datasets in the historical sub-time period.

[0051] It is understandable that, based on the target centroid information corresponding to multiple high-dimensional datasets of the battery pack in any sub-time period, the first centroid spacing corresponding to each high-dimensional dataset in any sub-time period is calculated. Similarly, based on the historical target centroid information of the battery pack in previous historical sub-time periods, the second centroid spacing corresponding to each high-dimensional dataset in previous historical sub-time periods is calculated. Based on the first centroid spacing corresponding to each high-dimensional dataset in any sub-time period and the second centroid spacing corresponding to each high-dimensional dataset in previous historical sub-time periods, the average centroid spacing corresponding to each high-dimensional dataset is determined by averaging. Combining the second centroid spacing of previous historical sub-time periods with the first centroid spacing of any sub-time period allows for the identification of the battery pack's state evolution trend over time, thereby identifying early signs of thermal runaway and improving the accuracy of determining the probability of battery pack thermal runaway.

[0052] In one optional embodiment, the target thermal runaway flag of the battery pack in any sub-time period is determined based on the average centroid spacing corresponding to multiple high-dimensional datasets and the first-order difference of the centroids corresponding to multiple high-dimensional datasets, including: obtaining a first thermal runaway flag based on the average centroid spacing corresponding to multiple high-dimensional datasets and a pre-set first threshold; obtaining a second thermal runaway flag based on the first-order difference of the centroids corresponding to multiple high-dimensional datasets and a pre-set second threshold; and determining the target thermal runaway flag based on the first thermal runaway flag and the second thermal runaway flag.

[0053] It can be understood that the average centroid spacing corresponding to multiple high-dimensional datasets is compared with a pre-set first threshold. If any of the average centroid spacings is greater than the first threshold, then the first thermal runaway flag is determined as indicating that the battery pack has a risk of thermal runaway in any sub-time period; if all the average centroid spacings are less than or equal to the first threshold, then the first thermal runaway flag is determined as indicating that the battery pack does not have a risk of thermal runaway in any sub-time period. Similarly, the first-order centroid differences corresponding to multiple high-dimensional datasets are compared with a pre-set second threshold. If any of the first-order centroid differences is greater than the second threshold, then the second thermal runaway flag is determined as indicating that the battery pack has a risk of thermal runaway in any sub-time period; if all the first-order centroid differences are less than or equal to the second threshold, then the second thermal runaway flag is determined as indicating that the battery pack does not have a risk of thermal runaway in any sub-time period. If either the first thermal runaway indicator or the second thermal runaway indicator is identified as indicating a risk of thermal runaway for the battery pack in any sub-time period, then the target thermal runaway indicator for that sub-time period is determined to indicate a risk of thermal runaway. Conversely, if both the first and second thermal runaway indicators indicate no risk of thermal runaway for the battery pack in any sub-time period, then the target thermal runaway indicator for that sub-time period is determined to indicate no risk of thermal runaway for the battery pack in any sub-time period. By combining the average centroid spacing and the first-order difference of the centroids to determine the target thermal runaway indicator for the battery pack, compared to judging with a single indicator, a more comprehensive assessment of the battery pack's thermal runaway state can be achieved, improving the accuracy of the battery pack's thermal runaway state determination results, and consequently, improving the accuracy of the battery pack's thermal runaway probability determination results.

[0054] Optionally, the probability of thermal runaway of the battery pack during a predetermined time period can be calculated as follows: for the target centroid information of the operating state i in time segment t, determine the average centroid spacing of its target centroid information. First difference from the center of mass like satisfy Then it is determined that the battery pack has a risk of thermal runaway in its operating state i at time segment t; if satisfy This indicates that the battery pack is at risk of thermal runaway in operating state i during time segment t.

[0055] Optionally, the above-described determination method can be used to determine the target thermal runaway indicator of the battery pack. First, it is determined whether there is an average centroid spacing greater than a first threshold among the average centroid spacings corresponding to the multiple operating states of the battery pack in time segment t. If so, the first thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack has a risk of thermal runaway in time segment t; otherwise, the first thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack does not have a risk of thermal runaway in time segment t. Second, it is determined whether there is a first-order centroid difference greater than a second threshold among the first-order centroid differences corresponding to the multiple operating states of the battery pack in time segment t. If so, the second thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack has a risk of thermal runaway in time segment t; otherwise, the second thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack does not have a risk of thermal runaway in time segment t. Finally, if either the first thermal runaway indicator or the second thermal runaway indicator is determined to indicate a risk of thermal runaway, then the target thermal runaway indicator for the battery pack at time segment t is determined to indicate a risk of thermal runaway; otherwise, the target thermal runaway indicator for the battery pack at time segment t is determined to indicate a lack of thermal runaway risk.

[0056] Step S106: Based on the target thermal runaway flags corresponding to multiple sub-time periods, determine the probability of thermal runaway of the battery pack in a predetermined time period. The target thermal runaway flags corresponding to multiple sub-time periods are obtained by determining the target thermal runaway flags for any sub-time period.

[0057] It is understandable that by identifying the target thermal runaway markers for the battery pack in any sub-time period, the target thermal runaway markers corresponding to the battery pack in multiple sub-time periods within a predetermined time period can be determined. Based on the target thermal runaway markers corresponding to these multiple sub-time periods within the predetermined time period, the probability of thermal runaway of the battery pack during the predetermined time period can be determined. Calculating the probability of thermal runaway of the battery pack during the predetermined time period using the target thermal runaway markers corresponding to these multiple sub-time periods provides a quantitative assessment of the battery pack's thermal runaway risk, improves the accuracy of the determined probability, and ultimately enables early warning and preventative measures for battery pack thermal runaway.

[0058] In one optional embodiment, the probability of thermal runaway of the battery pack during a predetermined time period is determined based on the target thermal runaway flags corresponding to multiple sub-time periods, including: determining a first number of target thermal runaway flags indicating a risk of thermal runaway among the target thermal runaway flags corresponding to the multiple sub-time periods, and a second number of the multiple sub-time periods included in the predetermined time period; and determining the thermal runaway probability based on the first number and the second number.

[0059] It is understood that a first number of target thermal runaway markers identified as having a risk of thermal runaway are determined among the target thermal runaway markers corresponding to multiple sub-time periods of the predetermined time period, and a second number are determined among the multiple sub-time periods included in the predetermined time period. The probability of thermal runaway of the battery pack within the predetermined time period is determined based on the ratio of the first number and the second number. By statistically analyzing the number of target thermal runaway markers identified as having a risk of thermal runaway within the predetermined time period and calculating the probability of thermal runaway of the battery pack within the predetermined time period, a quantitative analysis of the thermal runaway risk of the battery pack can be achieved, improving the accuracy of the determination results of the thermal runaway probability of the battery pack.

[0060] Optionally, if the battery pack has a risk of thermal runaway during a time segment, the target thermal runaway flag for that time segment is recorded as 1; otherwise, it is recorded as 0. Using the above method, a first number of target thermal runaway flags of the battery pack being 1 within a predetermined time period including multiple time segments, and a second number of time segments within the predetermined time period are determined. The ratio of the first number to the second number is the probability of thermal runaway of the battery pack during the predetermined time period.

[0061] Optionally, thermal runaway fault warning for the battery pack can be performed as follows: The increasing and decreasing trends of the offset of coordinate points in two-dimensional space are analyzed using first-order centroid difference and threshold analysis, and it is further analyzed whether the increasing and decreasing trends of the offset of coordinate points in two-dimensional space gradually expand before thermal runaway. Based on the above analysis results, a thermal runaway fault warning for the battery pack is performed. After completing the thermal runaway fault warning for the battery pack, the above threshold is learned and updated, laying the foundation for subsequent thermal runaway fault warnings for the battery pack.

[0062] Through the above steps S102 to S106, the goal of collecting and analyzing the operating data of the battery pack during a predetermined time period can be achieved, determining the target thermal runaway risk indicators corresponding to the multiple sub-time periods included in the predetermined time period, and thus obtaining the thermal runaway probability of the battery pack within the predetermined time period. This achieves the technical effect of improving the accuracy of the determination result of the thermal runaway probability of the battery pack, thereby solving the technical problem of low accuracy in the determination result of the thermal runaway probability of the battery pack in related technologies.

[0063] Based on the above embodiments and optional embodiments, this application proposes an implementation method for determining the probability of thermal runaway of a battery pack, which can be understood as a thermal runaway identification method based on high-dimensional spatial hierarchical center embedding. Figure 2 This is a first flowchart of an optional method for determining the thermal runaway probability of a battery pack according to an embodiment of this application, as shown below. Figure 2 As shown, the steps of this method include:

[0064] Step S1 involves uploading the battery pack operating data from the battery management system, including voltage, current, insulation resistance, and temperature, to a big data cloud platform in real time. After data preprocessing, time-series data of physical quantities such as current and insulation resistance are obtained. The overall operating data of the battery pack is then segmented to obtain time-series data of voltage and temperature. The aforementioned time-series data of physical quantities such as current and insulation resistance, as well as the time-series data of voltage and temperature, constitute the preprocessed standard data of the battery pack.

[0065] Operational data such as voltage, current, insulation resistance, and temperature of the electric vehicle battery pack are obtained from a big data cloud platform, with a sampling interval of 30 seconds. This operational data undergoes preprocessing, including removing outliers, filling in a small number of missing data points, and separating the voltage and temperature of all battery cells in the battery pack, resulting in continuous time-series data of the individual cell voltages, temperatures, currents, and insulation resistance of the battery pack.

[0066] Taking voltage as an example, this section explains the method for determining continuous time series data for voltage. Voltage data with intervals exceeding 300 seconds are divided into two different time series datasets. Within any given time series dataset, for a small number of voltage data points that are 0, interpolation is used to fill in the missing values, and a small number of noisy data points exceeding a specific range are removed. Finally, the remaining voltage data are organized into a time series to form continuous time series data for voltage.

[0067] Step S2: The standard data of charging state, discharging state and post-charging resting state in the standard data of the preprocessed battery pack obtained in step S1 are identified and tagged to obtain the tagged operation data of the battery pack in a time segment (i.e. any sub-time period), such as 1 day (i.e., the standard data corresponding to multiple operation states respectively).

[0068] The preprocessed standard data of the battery pack is segmented according to three state judgment logics (i.e., operating state judgment logic), and the standard data of the charging state, discharging state, and post-charging resting state of the battery pack in the time segment are extracted. The standard data of the charging state, discharging state, and post-charging resting state are then labeled (Charge, corresponding to the charging state; Discharge, corresponding to the discharging state; Charge_still, corresponding to the post-charging resting state).

[0069] The charging status judgment logic is as follows: when the electric vehicle is in a stationary state, the current is greater than 0, the charging / discharging flag is 1, and the State of Health (SOH) gradually increases; when the discharging state judgment logic is as the electric vehicle's speed changes continuously, the SOH gradually decreases, the current is generally less than 0, and the charging / discharging flag is 0; when the electric vehicle is in a stationary state after charging, the electric vehicle's speed is 0, the SOH starts to gradually increase, then reaches a stable state, and the current is greater than 0.

[0070] Step S3: Extract features from the labeled operating data of the battery pack in each time segment obtained in step S2, calculate the statistical features of the labeled operating data of the battery pack in each time segment, such as mean and variance, and determine the high-dimensional dataset.

[0071] The statistical characteristics of the tagged operational data of the battery pack in each time segment, combined with the continuous time series data obtained in step S1, form a high-dimensional dataset of the battery pack in that time segment. The size of the high-dimensional dataset can be dynamically changed based on the size of the battery pack.

[0072] Step S4: Cluster the high-dimensional datasets corresponding to the multiple operating states of the battery pack in the time segment obtained in step S3, and the number of clusters is 2.

[0073] Clustering was performed on the high-dimensional datasets corresponding to multiple operating states of the battery pack during different time segments. This is because thermal runaway of the battery pack generally occurs in two states: occurring and not occurring, with completely different operational data structures. Furthermore, before thermal runaway occurs in an electric vehicle, the operational data structure of the battery pack undergoes a slight shift towards the thermal runaway operational data structure. Based on the above analysis, the cluster N of the high-dimensional dataset was set to 2. Table 1 shows the high-dimensional centroids obtained after clustering a certain high-dimensional dataset.

[0074] Table 1 shows the high-dimensional centroids obtained after clustering a certain high-dimensional dataset.

[0075] "Occurrence" cluster centroid 3881 3878 3877 3878 3876 …… "Not occurring" cluster centroid 3612 3617 3617 3618 3617 ……

[0076] Step S5: Through the clustering in step S4, the centroids corresponding to multiple high-dimensional datasets that correspond one-to-one with multiple running states are obtained, and information is extracted from the centroids.

[0077] Based on the clustering results above, centroids corresponding to multiple high-dimensional datasets that are one-to-one with multiple running states are obtained. These centroids are high-dimensional centroids. Different methods are used to extract information from the high-dimensional centroids, including but not limited to t-distribution random neighborhood embedding and singular value decomposition, to extract multiple 2D information of the centroids (i.e., multiple initial centroid information). After fusion, the main information of the centroids (i.e., target centroid information) is obtained. Figure 3This is a second flowchart of an optional method for determining the thermal runaway probability of a battery pack according to an embodiment of this application, as shown below. Figure 3 The diagram illustrates the process of extracting and fusing information from the centroids of a high-dimensional dataset to obtain the target centroid information. For a high-dimensional dataset X, it is first clustered to obtain its centroids. Information is then extracted from the centroids of X using both t-distributed random neighborhood embedding and singular value decomposition methods, yielding two initial centroid information sets. These two initial centroid information sets are then normalized and fused to obtain the target centroid information for the high-dimensional dataset X.

[0078] Step S6: Normalize the information extraction results of step S5 to obtain target centroid information corresponding to multiple operating states. Perform multi-logic early warning for thermal runaway of battery pack based on the normalized target centroid information.

[0079] The above method is used to obtain the target centroid information corresponding to multiple operating states of the battery pack in multiple time segments, and then normalizes it. The obtained normalized two-dimensional target centroid information is mapped to a two-dimensional space by day (i.e., by time segment). By analyzing the two-dimensional target centroid information of the battery pack in different time segments, thermal runaway early warning of the battery pack is provided.

[0080] Multiple initial centroid information is obtained based on the information extraction results. This initial centroid information is then normalized and fused to obtain two-dimensional target centroid information. This two-dimensional target centroid information is mapped onto a two-dimensional space, forming two coordinate points (one corresponding to the centroid of the thermal runaway cluster, and the other to the centroid of the cluster that has not experienced thermal runaway; the distance between the two coordinate points is the centroid spacing of the battery pack in that time segment). The above method is used to obtain two coordinate points of the battery pack in two-dimensional space at different time segments. By analyzing these two coordinate points of the battery pack in two-dimensional space at different time segments, thermal runaway fault warnings are provided for the battery pack, and the probability of thermal runaway of the battery pack within a predetermined time period is determined.

[0081] Thermal runaway fault warning for the battery pack is achieved through the following method: First-order centroid difference and threshold analysis are used to analyze the rising and falling trends of the offset of coordinate points in two-dimensional space, and to analyze whether the trend of these offsets gradually increases before thermal runaway. Based on the analysis results, a thermal runaway fault warning is generated for the battery pack. After completing the thermal runaway fault warning, the threshold is learned and updated to lay the foundation for subsequent thermal runaway fault warnings for the battery pack.

[0082] The probability of thermal runaway of the battery pack within a predetermined time period is calculated as follows: For the target centroid information of operating state i in time segment t, the average centroid spacing of its target centroid information is determined. First difference from the center of mass like satisfy Then it is determined that the battery pack has a risk of thermal runaway in its operating state i at time segment t; if satisfy This indicates that the battery pack is at risk of thermal runaway in operating state i during time segment t.

[0083] Using the above determination method, firstly, it is determined whether there is an average centroid spacing greater than a first threshold among the average centroid spacings corresponding to the multiple operating states of the battery pack in time segment t. If so, the first thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack has a risk of thermal runaway in time segment t; otherwise, the first thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack does not have a risk of thermal runaway in time segment t. Secondly, it is determined whether there is a first-order centroid difference greater than a second threshold among the first-order centroid differences corresponding to the multiple operating states of the battery pack in time segment t. If so, the second thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack has a risk of thermal runaway in time segment t; otherwise, the second thermal runaway indicator of the battery pack in time segment t is determined to indicate that the battery pack does not have a risk of thermal runaway in time segment t. Finally, if either the first thermal runaway indicator or the second thermal runaway indicator is determined to indicate a risk of thermal runaway, then the target thermal runaway indicator for the battery pack at time segment t is determined to indicate a risk of thermal runaway; otherwise, the target thermal runaway indicator for the battery pack at time segment t is determined to indicate a lack of thermal runaway risk.

[0084] If the battery pack is at risk of thermal runaway during a time segment, the target thermal runaway flag for that time segment is recorded as 1; otherwise, it is recorded as 0. Using the above method, a first number of target thermal runaway flags for the battery pack being 1 within a predetermined time period, including multiple time segments, and a second number of time segments within the predetermined time period are determined. The ratio of the first number to the second number is the probability of thermal runaway of the battery pack during the predetermined time period.

[0085] The above-mentioned optional implementation methods achieve at least the following effects: By preprocessing the operational data, the quality of the operational data can be improved, and the impact of data noise on the determination of the thermal runaway probability can be reduced. At the same time, by segmenting the operational data according to the operational state, the characteristics of the battery pack under different operational states can be identified more accurately, providing a rich state perspective for the determination of the thermal runaway probability of the battery pack. By fusing information from multiple initial centroids, the uncertainty caused by random fluctuations in individual initial centroid information can be reduced, improving the accuracy of the determination of the thermal runaway probability of the battery pack. By combining the average centroid spacing and the first-order difference of the centroids to determine the target thermal runaway marker of the battery pack, compared with the judgment of a single index, the thermal runaway state of the battery pack can be evaluated more comprehensively, improving the accuracy of the determination of the thermal runaway state of the battery pack, and thus improving the accuracy of the determination of the thermal runaway probability of the battery pack.

[0086] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0087] This embodiment also provides a device for determining the probability of thermal runaway of a battery pack. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the terms "module" and "device" can refer to a combination of software and / or hardware that performs a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementations, or a combination of software and hardware, are also possible and contemplated.

[0088] According to an embodiment of this application, an apparatus embodiment for implementing a method for determining the probability of thermal runaway of a battery pack is also provided. Figure 4 This is a schematic diagram of a battery pack thermal runaway probability determination device according to an embodiment of this application, such as... Figure 4 As shown, the above-mentioned thermal runaway probability determination device for battery pack includes a data acquisition module 402, a target thermal runaway flag determination module 404, and a thermal runaway probability determination module 406. The device will be described below.

[0089] The data acquisition module 402 is used to collect multiple sets of operating data of the battery pack during a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period.

[0090] The target thermal runaway flag determination module 404 is connected to the data acquisition module 402 and is used to determine the target thermal runaway flag of the battery pack in any sub-time period based on the operating data of any sub-time period. The target thermal runaway flag is used to indicate whether there is a risk of thermal runaway in the battery pack in any sub-time period.

[0091] The thermal runaway probability determination module 406 is connected to the target thermal runaway flag determination module 404. It is used to determine the thermal runaway probability of the battery pack in a predetermined time period based on the target thermal runaway flags corresponding to multiple sub-time periods. The target thermal runaway flags corresponding to multiple sub-time periods are obtained by determining the target thermal runaway flag of any sub-time period.

[0092] In a battery pack thermal runaway probability determination device provided in this application embodiment, a data acquisition module 402 is set up to collect multiple sets of operating data of the battery pack during a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period; a target thermal runaway flag determination module 404 is connected to the data acquisition module 402, and is used to determine the target thermal runaway flag of the battery pack in any sub-time period based on the operating data of any sub-time period, wherein the target thermal runaway flag is used to indicate whether there is a risk of thermal runaway of the battery pack in any sub-time period; a thermal runaway probability determination module 406 is connected to the target thermal runaway flag determination module 404, and is used to determine the thermal runaway probability of the battery pack during the predetermined time period based on the target thermal runaway flags corresponding to the multiple sub-time periods respectively, wherein the target thermal runaway flags corresponding to the multiple sub-time periods are obtained by determining the target thermal runaway flag of any sub-time period. The goal is to collect and analyze the operating data of the battery pack over a predetermined time period, determine the target thermal runaway risk indicators corresponding to multiple sub-time periods within the predetermined time period, and thus obtain the thermal runaway probability of the battery pack within the predetermined time period. This achieves the technical effect of improving the accuracy of the determination result of the thermal runaway probability of the battery pack, thereby solving the technical problem of low accuracy in the determination result of the thermal runaway probability of the battery pack in related technologies.

[0093] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0094] It should be noted that the data acquisition module 402, the target thermal runaway flag determination module 404, and the thermal runaway probability determination module 406 mentioned above correspond to steps S102 to S106 in the embodiments. The instances and application scenarios implemented by the above modules and their corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run on a computer terminal.

[0095] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the embodiments, and will not be repeated here.

[0096] The aforementioned battery pack thermal runaway probability determination device may further include a processor and a memory. The data acquisition module 402, the target thermal runaway flag determination module 404, and the thermal runaway probability determination module 406 are all stored in the memory as program units, and the processor executes the aforementioned program units stored in the memory to realize the corresponding functions.

[0097] The processor contains a core that retrieves the corresponding program unit from memory. One or more cores may be configured. Memory may include non-persistent memory in computer-readable media, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory includes at least one memory chip.

[0098] This application provides a non-volatile storage medium storing a program that, when executed by a processor, implements a method for determining the probability of thermal runaway of a battery pack.

[0099] This application provides an electronic device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: collecting multiple sets of operational data from a battery pack over a predetermined time period, wherein each set of operational data corresponds one-to-one with multiple sub-time periods included in the predetermined time period; for any sub-time period, determining a target thermal runaway flag for the battery pack in that sub-time period based on the operational data of that sub-time period, wherein the target thermal runaway flag indicates whether there is a risk of thermal runaway in the battery pack in that sub-time period; and determining the probability of thermal runaway of the battery pack over the predetermined time period based on the target thermal runaway flags corresponding to the multiple sub-time periods, wherein the target thermal runaway flags corresponding to the multiple sub-time periods are obtained by determining the target thermal runaway flag for any sub-time period. The device described herein may be a server, PC, etc.

[0100] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program with the following method steps: collecting multiple sets of operating data of a battery pack over a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period; for any sub-time period, determining a target thermal runaway flag of the battery pack in that sub-time period based on the operating data of that sub-time period, wherein the target thermal runaway flag is used to indicate whether there is a risk of thermal runaway in the battery pack in that sub-time period; and determining the probability of thermal runaway of the battery pack in the predetermined time period based on the target thermal runaway flags corresponding to the multiple sub-time periods respectively, wherein the target thermal runaway flags corresponding to the multiple sub-time periods are obtained by determining the target thermal runaway flag of any sub-time period.

[0101] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0103] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0104] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0105] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0106] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0107] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0108] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0109] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for determining the probability of thermal runaway in a battery pack, characterized in that, include: Collect multiple sets of operating data of the battery pack during a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period; For any one of the plurality of sub-time periods, based on the operating data of any one sub-time period, a target thermal runaway flag for the battery pack is determined in that sub-time period, wherein the target thermal runaway flag is used to indicate whether the battery pack is at risk of thermal runaway in that sub-time period; Based on the target thermal runaway flags corresponding to the plurality of sub-time periods, the probability of thermal runaway of the battery pack in the predetermined time period is determined, wherein the target thermal runaway flags corresponding to the plurality of sub-time periods are obtained by determining the target thermal runaway flags of any sub-time period. The step of determining the thermal runaway probability of the battery pack in the predetermined time period based on the target thermal runaway flags corresponding to the plurality of sub-time periods includes: determining a first number of target thermal runaway flags indicating a risk of thermal runaway among the target thermal runaway flags corresponding to the plurality of sub-time periods, and a second number of the plurality of sub-time periods included in the predetermined time period; and determining the thermal runaway probability based on the first number and the second number.

2. The method according to claim 1, characterized in that, The determination of the target thermal runaway indicator of the battery pack in any sub-time period based on the operating data of any sub-time period includes: Based on the operational data, determine the high-dimensional datasets corresponding to the multiple operational states of the battery pack in any sub-time period; Determine the target centroid information corresponding to multiple high-dimensional datasets for any sub-time period, wherein the multiple high-dimensional datasets correspond one-to-one with the multiple running states; Based on the target centroid information corresponding to the multiple high-dimensional datasets, the target thermal runaway indicator of the battery pack is determined in any sub-time period.

3. The method according to claim 2, characterized in that, The step of determining the high-dimensional datasets corresponding to multiple operating states of the battery pack in any sub-time period based on the operational data includes: The operational data is preprocessed to obtain standard data of the battery pack in any sub-time period, wherein the preprocessing includes at least one of the following: data cleaning, data filling and data splitting; The standard data is segmented according to the operating state judgment logic to obtain standard data corresponding to multiple operating states. The operating state judgment logic is used to identify and classify the data of the battery pack under different operating states. Based on the standard data corresponding to the multiple operating states, the statistical characteristics corresponding to the multiple operating states are determined. Based on the standard data corresponding to the multiple operating states and the statistical characteristics corresponding to the multiple operating states, the high-dimensional datasets corresponding to the multiple operating states are determined.

4. The method according to claim 2, characterized in that, The determination of the target centroid information corresponding to the multiple high-dimensional datasets for any sub-time period includes: For any running state among multiple running states, clustering is performed on any state high-dimensional dataset of any running state to obtain any state centroid of the high-dimensional dataset of any running state. Information is extracted from the centroid of any state to obtain multiple initial centroid information for any state centroid; By fusing the multiple initial centroid information, the target centroid information of any state in any state high-dimensional dataset is obtained; By determining the target centroid information for any of the states, the target centroid information corresponding to each of the multiple operating states is obtained.

5. The method according to claim 2, characterized in that, The step of determining the target thermal runaway indicator of the battery pack in any sub-time period based on the target centroid information corresponding to the multiple high-dimensional datasets includes: Based on the target centroid information corresponding to the multiple high-dimensional datasets, the average centroid spacing and the first-order centroid difference corresponding to the multiple high-dimensional datasets are determined. The average centroid spacing represents the average distance between the data in the high-dimensional dataset and the centroid, and the first-order centroid difference is used to represent the change of the centroid over time. Based on the average centroid spacing corresponding to the multiple high-dimensional datasets and the first-order difference of the centroids corresponding to the multiple high-dimensional datasets, the target thermal runaway indicator of the battery pack in any sub-time period is determined.

6. The method according to claim 5, characterized in that, The step of determining the average centroid spacing corresponding to each of the multiple high-dimensional datasets based on the target centroid information of each dataset includes: Based on the target centroid information corresponding to the multiple high-dimensional datasets respectively, the first centroid spacing corresponding to the multiple high-dimensional datasets for any sub-time period is determined; Based on the historical target centroid information of the historical sub-time periods preceding any of the aforementioned sub-time periods, the second centroid spacing corresponding to the battery pack in multiple high-dimensional datasets of the historical sub-time periods is determined. Based on the first centroid spacing corresponding to the multiple high-dimensional datasets in any sub-time period, and the second centroid spacing corresponding to the multiple high-dimensional datasets in the historical sub-time period, the average centroid spacing corresponding to the multiple high-dimensional datasets is determined.

7. The method according to claim 5, characterized in that, The step of determining the target thermal runaway indicator of the battery pack in any sub-time period based on the average centroid spacing corresponding to the multiple high-dimensional datasets and the first-order difference of the centroids corresponding to the multiple high-dimensional datasets includes: Based on the average centroid spacing corresponding to the multiple high-dimensional datasets and the preset first threshold, a first thermal runaway indicator is obtained; Based on the first-order difference of the centroid corresponding to the multiple high-dimensional datasets, and the pre-set second threshold, the second thermal runaway flag is obtained; The target thermal runaway flag is determined based on the first thermal runaway flag and the second thermal runaway flag.

8. A device for determining the probability of thermal runaway in a battery pack, characterized in that, include: The data acquisition module is used to collect multiple sets of operating data of the battery pack during a predetermined time period, wherein the multiple sets of operating data correspond one-to-one with multiple sub-time periods included in the predetermined time period; The target thermal runaway flag determination module is used to determine the target thermal runaway flag of the battery pack in any one of the plurality of sub-time periods based on the operating data of the any one sub-time period, wherein the target thermal runaway flag is used to indicate whether the battery pack has a risk of thermal runaway in the any one sub-time period; The thermal runaway probability determination module is used to determine the thermal runaway probability of the battery pack in the predetermined time period based on the target thermal runaway flags corresponding to the plurality of sub-time periods respectively, wherein the target thermal runaway flags corresponding to the plurality of sub-time periods are obtained by determining the target thermal runaway flags of any sub-time period; The thermal runaway probability determination module is further configured to: determine a first number of target thermal runaway markers representing the risk of thermal runaway among the target thermal runaway markers corresponding to the plurality of sub-time periods, and a second number of the plurality of sub-time periods included in the predetermined time period; and determine the thermal runaway probability based on the first number and the second number.

9. An electronic device, characterized in that, include: One or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method for determining the thermal runaway probability of a battery pack according to any one of claims 1 to 7.

Images