Power battery failure early warning method and device, electronic equipment and storage medium

Abstract

The application relates to the field of battery safety technology and provides a power battery fault early warning method and device, electronic equipment and a storage medium, the method comprises the following steps: collecting multiple data information of a power battery at equal intervals within a preset time length, the power battery comprises multiple battery packs, each data information comprises a first highest single battery temperature corresponding to the multiple battery packs, a total current and a residual capacity of the power battery, the multiple data information correspond to multiple collection periods one by one; multiple charging and discharging sections of the power battery within the preset time length are acquired, and a second highest single battery temperature corresponding to each charging and discharging section is determined from the multiple first highest single battery temperatures; according to a statistical value calculation result of each second highest single battery temperature, sub-temperatures included in the second highest single battery temperature are classified and marked to obtain temperature characteristic data, and the sub-temperatures belong to the multiple first highest single battery temperatures; and according to the multiple temperature characteristic data, the multiple total currents and the multiple residual capacities, a fault early warning result of the power battery is determined. The method realizes real-time diagnosis based on vehicle data of different charging and discharging sections, positions an abnormal temperature in advance and gives an early warning, and effectively reduces the fault occurrence rate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery safety, and in particular to a fault early warning method and device for a power battery, an electronic device and a storage medium. BACKGROUND

[0002] Under the dual pressures of environmental pollution and energy crisis, countries around the world have begun to pay high attention to new energy, and the advantages of electric equipment (such as electric vehicles, electric heavy trucks, etc.) have been recognized. Unlike traditional fuel equipment, electric equipment uses power batteries as the energy source of the whole vehicle. A power battery is usually composed of thousands of lithium ion batteries connected in series and in parallel to meet the power demand of electric equipment. However, as the power battery ages and is misused, the battery pack included in the power battery may still fail, and even in extreme cases, it may cause a catastrophic accident, which has a significant impact on the safety of the power battery. The main faults of the power battery include battery high-temperature fault, System on Chip (SoC) low alarm and poor consistency of the power battery. Among them, the battery high-temperature fault is a fault that often occurs in electric equipment. During the operation of the power battery, the thermal effect will affect the temperature and electrochemical performance of the power battery, such as charging and discharging performance, resistance, etc., resulting in a battery high-temperature fault. In addition, the temperature rise of the power battery and the heat accumulated inside may cause thermal runaway, and even in extreme cases, it may cause the equipment to burn or explode, causing economic losses and poor driving experience for users.

[0003] Traditional fault diagnosis methods often seek the physical relationship between fault factors, and generally only when obvious fault phenomena occur, diagnosis will be performed. However, for a complex power battery, it is difficult to determine the potential fault and the degree of fault occurrence, and there is difficulty in real-time online fault diagnosis, and there is a lack of foresight in fault judgment. SUMMARY

[0004] The present application provides a fault early warning method and device for a power battery, an electronic device and a storage medium, to solve the defects that the prior art is difficult to determine the potential fault and the degree of fault occurrence of the power battery, and there is difficulty in real-time online fault diagnosis, and there is a lack of foresight in fault judgment. By taking the whole vehicle data (such as temperature characteristic data, total current and remaining power) of the power battery at different charging and discharging stages as the basis, real-time diagnosis of the power battery is realized. During the diagnosis process, the occurrence time point of the abnormal temperature can be accurately and in advance located, the fault is warned, the fault occurrence rate is effectively reduced, the battery service life is effectively prolonged, and accidents of the electric equipment due to battery failure during driving are avoided.

[0005] The present application provides a fault early warning method for a power battery, comprising the following steps.

[0006] Collect multiple data information of the power battery at equal intervals within a preset time length, the power battery includes multiple battery packs, each data information includes a first highest single cell temperature corresponding to the multiple battery packs, a total current and a remaining power of the power battery, the multiple data information correspond to multiple collection periods one by one; multiple charging and discharging sections of the power battery within the preset time length are obtained, and a second highest single cell temperature corresponding to each charging and discharging section is determined from the multiple first highest single cell temperatures; a sub-temperature included in the second highest single cell temperature is classified and marked according to a statistical value calculation result of each second highest single cell temperature, to obtain temperature feature data, the sub-temperature belongs to the multiple first highest single cell temperatures; and a fault warning result of the power battery is determined according to multiple temperature feature data, multiple total currents and multiple remaining powers.

[0007] According to the power battery fault warning method provided by the application, the statistical value calculation result of each second highest single cell temperature is obtained, and the second highest single cell temperature is classified and marked to obtain the temperature feature data corresponding to each of the multiple charging and discharging sections, including: traversing the multiple charging and discharging sections, for the current charging and discharging section, in the case that the current charging and discharging section is the first charging and discharging section, determining a first temperature mean value and a first maximum highest single cell temperature according to the second highest single cell temperature corresponding to the first charging and discharging section, and determining the first temperature mean value and the first maximum highest single cell temperature as the statistical value calculation result of the second highest single cell temperature; in the case that the current charging and discharging section is other charging and discharging section except the first charging and discharging section, determining a second temperature mean value, a second maximum highest single cell temperature, a temperature mean value difference and a temperature maximum value difference according to the second highest single cell temperature corresponding to the other charging and discharging section, and determining the second temperature mean value, the second maximum highest single cell temperature, the temperature mean value difference and the temperature maximum value difference as the statistical value calculation result of the second highest single cell temperature; and the corresponding second highest single cell temperature is classified and marked according to multiple statistical value calculation results, to obtain the temperature feature data corresponding to each of the multiple charging and discharging sections.

[0008] According to the power battery fault warning method provided by the application, the statistical value calculation result of each second highest single cell temperature is obtained, and the second highest single cell temperature is classified and marked to obtain the temperature feature data corresponding to each of the multiple charging and discharging sections, including: the following operations are performed for the statistical value calculation result of each second highest single cell temperature: in the case that the statistical value calculation result is greater than a preset fault threshold, the second highest single cell temperature is marked as abnormal, and abnormal state data is determined as the temperature feature data; and in the case that the statistical value calculation result is less than or equal to the preset fault threshold, the second highest single cell temperature is marked as normal, and normal state data is determined as the temperature feature data.

[0009] The method for early warning of the failure of the power battery according to the application comprises the following steps.

[0010] The method for early warning of the failure of the power battery according to the application comprises the following steps.

[0011] The method for early warning of the failure of the power battery according to the application comprises the following steps.

[0012] The method for early warning of the failure of the power battery according to the application comprises the following steps.

[0013] The application further provides a device for early warning of the failure of a power battery.

[0014] The data acquisition module is configured to acquire multiple data information of the power battery at equal intervals within a preset time length, wherein the power battery comprises multiple battery packs, each of the data information comprises a first highest single battery temperature corresponding to the multiple battery packs, a total current and a remaining capacity of the power battery, and the multiple data information correspond to multiple acquisition periods one by one.

[0015] The data processing module is configured to determine second highest cell temperatures corresponding to each charging and discharging section from the plurality of first highest cell temperatures, and to classify and mark sub-temperatures included in the second highest cell temperatures according to a calculation result of a statistical value of the second highest cell temperatures, to obtain temperature feature data, wherein the sub-temperatures belong to the plurality of first highest cell temperatures.

[0016] The temperature warning module is configured to determine a failure warning result of the power battery according to the plurality of temperature feature data, the plurality of total currents and the plurality of residual electric quantities.

[0017] The application further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the failure warning method of the power battery according to any one of the above.

[0018] The application further provides a non-transitory computer-readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the failure warning method of the power battery according to any one of the above.

[0019] The application further provides a computer program product, which includes a computer program, and the computer program is executable on a processor to implement the failure warning method of the power battery according to any one of the above.

[0020] The application provides a failure warning method, device, electronic device and storage medium of a power battery, which realizes real-time diagnosis of the power battery according to vehicle data (such as temperature feature data, total current and residual electric quantity) of the power battery in different charging and discharging sections, and can accurately and in advance locate the time point of abnormal temperature occurrence in the diagnosis process, and warn of failure, so as to effectively reduce the failure rate and effectively prolong the service life of the battery, and further avoid accidents of the electric device due to battery failure during driving. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0022] Figure 1 is a flowchart of the failure warning method of the power battery provided by the present application.

[0023] Figure 2a is a flowchart of data marking provided by the present application.

[0024] Figure 2bA construction flowchart of the LGBM model provided by the present application.

[0025] Figure 3 A structural schematic diagram of the fault early warning device of the power battery provided by the present application.

[0026] Figure 4 A structural schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0028] It should be noted that the execution subject involved in the embodiments of the present application can be the fault early warning device of the power battery, or the electronic device. Optionally, the electronic device can include a computer, a mobile terminal, a wearable device and the like.

[0029] The embodiments of the present application will be further described below taking the electronic device as an example.

[0030] Figure 1 A flowchart of the fault early warning method of the power battery provided by the present application. As shown in the figure, the method can include the following steps 101-104. Figure 1

[0031] Step 101, a plurality of first maximum single cell temperatures of the power battery are collected at equal intervals within a preset time length, and a plurality of charge and discharge sections of the power battery within the preset time length are obtained, as well as the total current and the remaining power of the power battery corresponding to each of the plurality of charge and discharge sections, the plurality of first maximum single cell temperatures corresponding to the plurality of collection periods one by one.

[0032] Among them, the preset time length is a time period set in advance, specifically, a time range from a starting time point to an ending time point.

[0033] The power battery is a battery used to provide electric energy for electric devices (such as electric vehicles, electric heavy trucks, etc.). The power battery is composed of multiple battery packs to provide sufficient electric energy and voltage. Generally, one battery pack is composed of multiple single cells, for example: a 350 heavy truck has 10 battery packs, each battery pack has 9 temperature probes, that is, one battery pack has 9 single cell temperatures, and the entire battery has 90 single cell temperatures.

[0034] ​The first highest single-cell temperature refers to the temperature corresponding to each of the multiple battery packs, i.e., the maximum value among multiple single-cell temperatures, used to evaluate the safety and performance of the power battery. The multiple first highest single-cell temperatures can be represented by T = [t1, ..., t...]. c ,…,t M ] indicates that M≥2, t c This represents the c-th highest monomer temperature among M highest monomer temperatures.

[0035] The charge and discharge phases typically refer to the charging and discharging stages that a power battery undergoes during operation. These two stages are crucial for battery performance, lifespan, and energy management.

[0036] Total current is the total current flowing through the power battery, used to determine the charging or discharging state of the power battery, as well as the load condition of the power battery.

[0037] Remaining charge refers to the electrical energy currently remaining in the battery, used to understand the battery's range and status. This remaining charge is usually expressed as a percentage or a specific charge value (such as kilowatt-hours).

[0038] The data acquisition period refers to the time interval between the acquisition of two adjacent data points. It should be noted that because data needs to be acquired at "equal intervals," this interval is fixed.

[0039] It should be noted that the aforementioned highest single-cell temperature, total current, and remaining charge can constitute the vehicle-wide data for the power battery.

[0040] During the acquisition of vehicle data, the electronic device can first determine a preset duration and acquisition cycle; then, within the preset duration and according to the acquisition cycle, it sequentially acquires the first highest cell temperature corresponding to the power battery. Since the first highest cell temperature can be acquired once within one acquisition cycle, the number of acquisition cycles is the same as the number of first highest cell temperatures, and multiple first highest cell temperatures correspond one-to-one with multiple acquisition cycles.

[0041] Since a power battery undergoes more than one charge-discharge cycle during operation, the electronic device can extract data from all charge-discharge segments of the power battery, thus extracting the total current corresponding to each segment. Specifically, the electronic device first resets the data index, then identifies data points with negative current, ensuring that the data points are continuous and at least ε, where ε ≥ 1. It then records the start and end index values ​​of the charging segment, while the discharging segment consists of data segments outside these index values. In this case, with n charging segments acquired, the index values ​​of these n charging segments can be represented as [[a0,a1],[a2,a3],…,[a…]. (n-1) ,a nrepresents, n≥2, at this time, the index values of the n discharge sections can be represented by [[a1, a2], [a3, a4], …, [a n (n+1) represents, based on this, the electronic device can obtain n charge and discharge sections, and the index values of the n discharge sections can be represented by [[a0, a2], [a2, a4], …, [a (n-1) (n+1) Finally, the electronic device extracts charge and discharge section data according to the index values of the n charge and discharge sections to obtain the total current corresponding to each charge and discharge section. At the same time, the electronic device can also obtain the residual capacity corresponding to each charge and discharge section.

[0042] In this way, the electronic device can determine vehicle data with high accuracy, so as to provide early warning for real high-temperature faults, and take preventive measures in advance to avoid potential safety hazards.

[0043] For example, assuming that the preset time length is 1 hour (h), that is, 3600 seconds (s), and the interval period is 10 s. The electronic device collects the first highest single cell temperature corresponding to the power battery once every 10 s within 2 h, based on which the electronic device can collect 360 first highest single cell temperatures within 1 h. Assuming that the lengths of the 6 charge and discharge sections within 1 h are the same, then each charge and discharge section corresponds to 60 total currents and 60 residual capacities.

[0044] Step 102, determining, from the plurality of first highest single cell temperatures, a second highest single cell temperature corresponding to each of the plurality of charge and discharge sections.

[0045] Since the electronic device may be in a certain charge and discharge section when collecting the first highest single cell temperature, after obtaining the plurality of first highest single cell temperatures, the electronic device can divide the plurality of first highest single cell temperatures to obtain a second highest single cell temperature corresponding to each of the plurality of charge and discharge sections, each second highest single cell temperature can include a plurality of sub-temperatures, and these sub-temperatures belong to the plurality of first highest single cell temperatures. In this way, the electronic device can more accurately understand the highest temperature performance of the power battery under different charge and discharge states by dividing the first highest single cell temperature according to different charge and discharge sections, and such refined management helps to improve the efficiency and safety of the power battery.

[0046] In some embodiments, the electronic device determines, from the plurality of first highest single cell temperatures, a second highest single cell temperature corresponding to each of the plurality of charge and discharge sections, which can include: the electronic device performs extreme value deletion and / or missing value processing on the plurality of first highest single cell temperatures to obtain a plurality of third highest single cell temperatures; and the electronic device determines, from the plurality of third highest single cell temperatures, a second highest single cell temperature corresponding to each of the plurality of charge and discharge sections.​​

[0047] In the temperature data collection process, the electronic device will inevitably have error data such as abnormal values and / or missing values. In order to avoid the influence of error data on the final fault warning result, the electronic device can perform data cleaning on the plurality of first highest single cell temperatures after obtaining the plurality of first highest single cell temperatures. Specifically, if an extreme value occurs, the extreme value is deleted, and if a missing value occurs, the missing value is deleted, to obtain a plurality of third highest single cell temperatures with high accuracy and reliability. The plurality of third highest single cell temperatures are the first highest single cell temperatures remaining after the error data is deleted. Then, the plurality of third highest single cell temperatures are used as the basis to divide into corresponding charge and discharge sections to obtain the second highest single cell temperature corresponding to each charge and discharge section, in order to determine the fault warning result with high accuracy.

[0048] Specifically, in the process of deleting extreme values from the plurality of first highest single cell temperatures, the plurality of first highest single cell temperatures can be compared one by one with a preset temperature range. If a certain first highest single cell temperature is not within the preset temperature range, it means that there is a problem with the temperature sensor, not the power battery itself. At this time, the first highest single cell temperature outside the preset temperature range can be determined as an extreme value, and the extreme value is deleted.

[0049] In the process of processing the plurality of first highest single cell temperatures for missing values, there may be missing sampling points in the data information, and the missing values have no effect on data analysis and battery fault diagnosis. Therefore, the missing values can be directly deleted to reduce the computational load of subsequent data analysis and improve data analysis efficiency.

[0050] The preset temperature range is a range composed of a first temperature threshold and a second temperature threshold, and the first temperature threshold is less than the second temperature threshold. Optionally, the preset temperature range can be set before the electronic device is shipped, or it can be customized by the user, which is not limited here.

[0051] Step 103, according to the statistical value calculation result of each second highest single cell temperature, classifying and marking each second highest single cell temperature to obtain temperature feature data corresponding to each charge and discharge section.

[0052] In order to improve the accuracy of the fault early warning result, after obtaining the second highest single cell temperature corresponding to each of the plurality of charging and discharging sections, the electronic device can calculate the second highest single cell temperature corresponding to each charging and discharging section to obtain a statistical value calculation result, and then classify and mark the second highest single cell temperature to obtain temperature feature data corresponding to the charging and discharging section. Based on this, the more charging and discharging sections there are, the more temperature feature data the electronic device will obtain. These temperature feature data are more representative and effective, and can more accurately reflect the temperature characteristics of the power battery under different charging and discharging states, so as to obtain a fault early warning result with higher accuracy.

[0053] In addition, the selection of the above temperature feature data is very important for using a neural network to perform a regression task, which helps the prediction model to capture the relationship between the data input and the fault output.

[0054] In some embodiments, the electronic device classifies and marks each second highest single cell temperature according to the statistical value calculation result of each second highest single cell temperature to obtain temperature feature data corresponding to each of the plurality of charging and discharging sections, which can include at least one of the following implementation manners.

[0055] Implementation manner 1: The electronic device traverses the plurality of charging and discharging sections, and for a current charging and discharging section, in a case where the current charging and discharging section is the first charging and discharging section, the electronic device determines a first temperature mean and a first maximum highest single cell temperature according to the second highest single cell temperature corresponding to the first charging and discharging section, and determines the first temperature mean and the first maximum highest single cell temperature as the statistical value calculation result of the second highest single cell temperature; in a case where the current charging and discharging section is a charging and discharging section other than the first charging and discharging section, the electronic device determines a second temperature mean, a second maximum highest single cell temperature, a temperature mean difference, and a temperature maximum difference according to the second highest single cell temperature corresponding to the other charging and discharging section, and determines the second temperature mean, the second maximum highest single cell temperature, the temperature mean difference, and the temperature maximum difference as the statistical value calculation result of the second highest single cell temperature; and the electronic device classifies and marks the corresponding second highest single cell temperature according to a plurality of statistical value calculation results to obtain temperature feature data corresponding to each of the plurality of charging and discharging sections.

[0056] The first temperature mean is used to reflect the central tendency of the second highest single cell temperature corresponding to the first charging and discharging section.

[0057] The first maximum highest single cell temperature is used to reflect the highest temperature reached by the power battery during the first charging and discharging process, so as to evaluate the safety of the power battery.

[0058] The second temperature mean is used to reflect the central tendency of the second highest single cell temperature corresponding to the other charging and discharging section.

[0059] The second maximum single-cell temperature is used to reflect the highest temperature reached by the power battery during other charging and discharging processes, in order to assess the safety of the power battery.

[0060] The temperature mean difference is used to monitor the temperature change between other charge / discharge stages and the previous charge / discharge stage.

[0061] The maximum temperature difference is used to monitor abnormal battery temperatures under other charge / discharge stages.

[0062] After acquiring the second highest single-cell temperature corresponding to each of the multiple charge / discharge segments, the electronic device can iterate through these multiple charge / discharge segments. For the current charge / discharge segment, it first determines whether the current charge / discharge segment is the first charge / discharge segment. If the current charge / discharge segment is the first charge / discharge segment, the second highest single-cell temperature corresponding to the first charge / discharge segment can be calculated to obtain the first average temperature and the first maximum single-cell temperature, and these are determined as the statistical value calculation result of the second highest single-cell temperature. If the current charge / discharge segment is not the first charge / discharge segment, it means that the current charge / discharge segment is another charge / discharge segment besides the first charge / discharge segment. At this time, the second highest single-cell temperature corresponding to the other charge / discharge segment can be calculated to obtain the second average temperature, the second maximum single-cell temperature, the temperature mean difference, and the temperature maximum difference, and these are determined as the statistical value calculation result of the second highest single-cell temperature. After traversing the multiple charging and discharging segments, the electronic device can obtain the statistical value calculation results corresponding to each of these charging and discharging segments. Then, it can classify and mark the corresponding second highest single-cell temperature to obtain the temperature characteristic data corresponding to each of these charging and discharging segments. This allows for the rapid identification of abnormal temperature changes, thereby enabling fault diagnosis and improving the efficiency of fault handling.

[0063] In some embodiments, the electronic device determines a first average temperature and a first maximum maximum single-cell temperature based on the second maximum single-cell temperature corresponding to the first charge / discharge segment. This may include: the electronic device acquiring a plurality of sub-temperatures included in the second maximum single-cell temperature corresponding to the first charge / discharge segment; the electronic device determining a first average temperature based on the values ​​of the plurality of sub-temperatures; and the electronic device determining the maximum temperature among the plurality of sub-temperatures as the first maximum maximum single-cell temperature.

[0064] Among them, multiple sub-temperatures belong to multiple first highest single-unit temperatures.

[0065] For example, suppose the second highest single-cell temperature corresponding to the first charge / discharge stage can be represented by T = [t1,…,t]. j ,…,t m ] represents, where m represents the number of sub-temperatures, M≥m≥2; t j This represents the value of the j-th sub-temperature out of m sub-temperatures. In this case, the average first temperature can be expressed by the formula... T_mean1 represents the first temperature mean value, represents the temperature and value; meanwhile, the electronic device can compare the m sub-temperatures and adopt T_max1 = max(t α ), to determine the maximum temperature t α as the first maximum highest single cell temperature T_max1, where t α represents the αth sub-temperature in the m sub-temperatures, α ∈ [1, …, m].

[0066] In some embodiments, the electronic device determines the temperature mean value difference and the temperature maximum value difference according to the second highest single cell temperature corresponding to the other charging and discharging section, which can include: the electronic device acquires a previous charging and discharging section of the other charging and discharging section, and a third temperature mean value and a third maximum highest single cell temperature corresponding to the previous charging and discharging section; the electronic device determines the temperature mean value difference according to the second temperature mean value and the third temperature mean value; and the electronic device determines the temperature maximum value difference according to the second maximum highest single cell temperature and the third maximum highest single cell temperature.

[0067] The third temperature mean value is used to reflect the central tendency of the second highest single cell temperature corresponding to the previous charging and discharging section.

[0068] The third maximum highest single cell temperature is used to reflect the highest temperature reached by the power battery in the previous charging and discharging process, so as to evaluate the safety of the power battery.

[0069] For example, assuming that the electronic device acquires W charging and discharging sections, the other charging and discharging section is the wth charging and discharging section, W ≥ 2, w ∈ [2, …, W], at this time, the temperature mean value difference can adopt the formula d_mean w = T_mean w - T_mean w-1 is calculated, where d_mean w represents the temperature mean value difference of the wth charging and discharging section; T_mean w represents the temperature mean value of the wth charging and discharging section, i.e., the second temperature mean value; T_mean w-1 represents the temperature mean value of the w-1th charging and discharging section, i.e., the temperature mean value of the previous charging and discharging section, i.e., the third temperature mean value; meanwhile, the temperature maximum value difference can adopt the formula d_max w = T_max w - T_max w-1 is calculated, where d_max w represents the temperature maximum value difference of the wth charging and discharging section; T_max w represents the second maximum highest single cell temperature; and T_max w-1 represents the third maximum highest single cell temperature.

[0070] It should be noted that the process of determining the third temperature mean value is similar to the process of determining the first temperature mean value, and the process of determining the third maximum highest single cell temperature is similar to the process of determining the first maximum highest single cell temperature, which will not be described in detail here.

[0071] Implementation 2: The following operations are performed on the statistical value calculation result of each second highest single cell temperature: in the case where the statistical value calculation result is greater than the preset failure threshold, the electronic device marks the second highest single cell temperature as abnormal, and determines the abnormal state data as the temperature feature data; in the case where the statistical value calculation result is less than or equal to the preset failure threshold, the electronic device marks the second highest single cell temperature as normal, and determines the normal state data as the temperature feature data.

[0072] Optionally, the preset failure threshold can be set before the electronic device is shipped, or can be customized by the user, which will not be limited here.

[0073] Exemplarily, Figure 2a is a flowchart of data marking provided by the present application. From Figure 2a As can be seen from the above, in the process of marking each second highest single cell temperature, the electronic device can first take one charging and discharging section as a sliding window according to the index value of the charging and discharging section; then set a preset failure threshold according to the business requirement, and compare the statistical value calculation result corresponding to the second highest single cell temperature with the preset failure threshold: if the statistical value calculation result is greater than the preset failure threshold, it means that the second highest single cell temperature is abnormal, at this time, the second highest single cell temperature can be marked as abnormal, and the abnormal state data can be determined as the temperature feature data; if the statistical value calculation result is less than or equal to the preset failure threshold, it means that the second highest single cell temperature is normal, at this time, the second highest single cell temperature can be marked as normal, and the normal state data can be determined as the temperature feature data. Based on this, all the temperature feature data obtained by the electronic device can be divided into two types of label feature data.

[0074] Optionally, in the case where the statistical value calculation result is greater than the preset failure threshold, the corresponding sliding window is marked as a first value; in the case where the statistical value calculation result is less than or equal to the preset failure threshold, the corresponding sliding window is marked as a second value.

[0075] Exemplarily, in the case where the first value is 1, the second value is 0; in the case where the first value is 0, the first value is 1.

[0076] It should be noted that in the case of the statistical value calculation result corresponding to the i th charging and discharging section, i ∈ [2, …, W], the statistical value calculation result includes the temperature mean value T_mean i , the temperature mean value difference T_maxi a maximum highest cell temperature d_mean i and a temperature maximum difference d_max i The respective corresponding sub-fault thresholds are different.

[0077] The preset fault threshold includes a first sub-fault threshold G1, a second sub-fault threshold G2, a third sub-fault threshold G3, and a fourth sub-fault threshold G4. At this time, the temperature mean T_mean i Corresponding to the first sub-fault threshold G1, the temperature mean difference T_max i Corresponding to the second sub-fault threshold G2, the maximum highest cell temperature d_mean i Corresponding to the third sub-fault threshold G3, the temperature maximum difference d_max i Corresponding to the fourth sub-fault threshold G4.

[0078] It can be understood that the statistical value calculation result greater than the preset fault threshold means that each data in the statistical value calculation result is greater than the respective corresponding sub-fault threshold.

[0079] Step 104, determining a fault warning result of the power battery according to the plurality of temperature characteristic data, the plurality of total currents, and the plurality of residual capacities.

[0080] After the electronic device obtains the temperature characteristic data, the total current, and the residual capacity corresponding to each of the plurality of charging and discharging sections, the electronic device can combine these three data to more accurately determine the health status and safety risk of the power battery, thereby giving a targeted fault warning result.

[0081] In some embodiments, the electronic device determines the fault warning result of the power battery according to the plurality of temperature characteristic data, the plurality of total currents, and the plurality of residual capacities, which can include: the electronic device inputs the plurality of temperature characteristic data, the plurality of total currents, and the plurality of residual capacities as input data to a Light Gradient Boosting Machine (LGBM) model to obtain the fault warning result of the power battery output by the LGBM model.

[0082] The LGBM model is a supervised model that needs to be labeled according to the characteristics of the data and business requirements in order to train and predict the subsequent model, thereby improving the accuracy and interpretability of data analysis and prediction. The LGBM model improves the prediction ability of the decision tree through gradient boosting, and the decision tree model involved can predict classification and continuous problems, such as predicting high-temperature faults of the battery. For the decision tree model, the key is to identify the optimal split point of the feature.

[0083] In the process of real-time fault detection of battery temperature by using the LGBM model, the electronic device can input multiple temperature feature data, multiple total currents and multiple residual capacities as input data into the LGBM model, analyze the data of each charging and discharging section through the LGBM model, and obtain a fault warning result of the power battery, which includes the current fault warning result of each charging and discharging section.

[0084] Specifically, in the process of training the LGBM model, the electronic device first acquires a data set of the power battery, which can be represented as {(x1, y1), …, (x β ,y β ), …, (x M ,y M )}, where x β represents the input data corresponding to the βth charging and discharging section, y β represents the current fault warning result corresponding to the βth charging and discharging section, x β ∈X, y β ∈Y, X represents the input data corresponding to all charging and discharging sections respectively, and Y represents the current fault warning result corresponding to all charging and discharging sections respectively.

[0085] Then, the LGBM model adopted by the electronic device is a linear combination with decision tree as the base function, which can be represented as f(x) = ∑p=1S h (x), where S represents the decision tree data corresponding to the LGBM model, S≥2; h p (x) represents the result of the pth decision tree in the S decision trees; f p (x; σ p ) represents the pth decision tree; and σ p represents the parameters of the pth decision tree.

[0086] Wherein, the parameter σ p can be calculated by the empirical risk minimization formula argmin σ ∈Ω L(y β , h p (x β )), where argmin(.) represents the index function; σ β represents the parameters of the βth decision tree, β∈[1,…,S]; L(y β , h p (x β )) represents the loss function of the pth decision tree; and h p (x β ) represents the fault warning result of the input data corresponding to the βth charging and discharging section under the pth decision tree.

[0087] Then, the electronic device determines the negative gradient, which can be calculated by the formula ∇f(x) = -∑p=1S ∇h (x), at this time, the output value that minimizes the loss function in the node region of the qth decision tree can be calculated by the formula h is calculated, wherein g q (x i ; σ p ) represents an output value; c represents a learning factor; b p-1 (x β ) represents a fault early warning result of the input data corresponding to the βth charge and discharge section under the (p-1)th decision tree; x β ∈r β，p represents x β corresponding to r β，p .

[0088] Finally, the electronic device accumulates the decision trees of each round of iteration to obtain a final learner, which can be represented as h , wherein h S (X) represents the accumulation result of the S decision trees.

[0089] It should be noted that the above learner is a trained LGBM model. The LGBM model finds the optimal split point based on the histogram theory, saves the time cost and memory overhead, and quickly obtains the current fault early warning result of each charge and discharge section.

[0090] Optionally, since the LGBM model has many model parameters, the value of the hyperparameters will affect the prediction accuracy, so a random search algorithm can be used to optimize the hyperparameters. The random search algorithm finds the best combination of hyperparameters by randomly sampling in the hyperparameter space, which greatly reduces the calculation amount of hyperparameter search and shortens the optimization time.

[0091] Optionally, in order to avoid overfitting and underfitting of the LGBM model and improve the prediction performance of the model, a K-fold cross-validation method can be used. The K-fold cross-validation method can divide the above data set into K equal parts in equal proportions, select one part as the test set, and use the remaining K-1 parts as the training set. The model is evaluated on the test set, and the evaluation index is recorded. Repeat the above steps to obtain K evaluation indexes. The average or weighted sum of the K evaluation indexes is obtained to obtain the target evaluation index corresponding to the LGBM model.

[0092] Exemplarily, Figure 2b is a construction flowchart of the LGBM model provided by the present application. From Figure 2bAs can be seen, the electronic device can use a random search algorithm to optimize the LGBM model, and combine the K-fold cross-validation method to evaluate the LGBM model. Specifically, the following steps are included: Step one, normalizing the data in the data set to generate a training set and a test set; Step two, initializing the model parameters of the LGBM model; Step three, using a random search algorithm to optimize the LGBM model to obtain the optimal hyperparameters of the LGBM model and update them; Step four, using the K-fold cross-validation method to evaluate the performance of the LGBM model; Step five, judging the prediction result of the LGBM model, if the prediction result is 1, an early warning is performed, and the early warning success rate is counted.

[0093] In the embodiment of the present application, a plurality of data information of the power battery is collected at equal intervals within a preset time period, the power battery includes a plurality of battery packs, each data information includes a first highest single temperature corresponding to the plurality of battery packs, a total current and a remaining power of the power battery, the plurality of data information corresponds to a plurality of collection periods one by one; a plurality of charge and discharge sections of the power battery within the preset time period are obtained, and a second highest single temperature corresponding to each charge and discharge section is determined from the plurality of first highest single temperatures; the sub-temperatures included in the second highest single temperature are classified and labeled according to the statistical value calculation result of each second highest single temperature, to obtain temperature feature data, the sub-temperatures belong to the plurality of first highest single temperatures; and a fault warning result of the power battery is determined according to the plurality of temperature feature data, the plurality of total currents and the plurality of remaining powers. The method realizes real-time diagnosis of the power battery based on the vehicle data (such as temperature feature data, total current and remaining power) of the power battery in different charge and discharge sections, and can accurately and in advance locate the time point of abnormal temperature occurrence during the diagnosis process, and warn of faults, so as to effectively reduce the fault occurrence rate and effectively prolong the service life of the battery, and further avoid accidents of the electric equipment due to battery failure during driving.

[0094] The power battery fault warning device provided by the present application is described below, and the power battery fault warning device described below can be correspondingly referred to the power battery fault warning method described above.

[0095] Figure 3 is a structural schematic diagram of the power battery fault warning device provided by the present application. As shown in Figure 3 , the device can include a data acquisition module 301, a data processing module 302 and a temperature warning module 303.

[0096] The data acquisition module 301 is configured to collect multiple data information of the power battery at equal intervals within a preset time length, the power battery includes multiple battery packs, each data information includes a first highest single cell temperature corresponding to the multiple battery packs, a total current and a remaining power of the power battery, and the multiple data information correspond to multiple collection periods one by one.

[0097] The data processing module 302 is configured to determine a second highest single cell temperature corresponding to each charging and discharging section from the multiple first highest single cell temperatures, and perform classification marking on a sub-temperature included in the second highest single cell temperature according to a statistical value calculation result of the second highest single cell temperature, to obtain temperature feature data, the sub-temperature belonging to the multiple first highest single cell temperatures.

[0098] The temperature warning module 303 is configured to determine a fault warning result of the power battery according to the multiple temperature feature data, the multiple total currents and the multiple remaining powers.

[0099] Optionally, the data processing module 302 is specifically configured to traverse the multiple charging and discharging sections, and for a current charging and discharging section, in a case that the current charging and discharging section is a first charging and discharging section, determine a first temperature mean value and a first maximum highest single cell temperature according to a second highest single cell temperature corresponding to the first charging and discharging section, and determine the first temperature mean value and the first maximum highest single cell temperature as a statistical value calculation result of the second highest single cell temperature; in a case that the current charging and discharging section is a charging and discharging section other than the first charging and discharging section, determine a second temperature mean value, a second maximum highest single cell temperature, a temperature mean value difference and a temperature maximum value difference according to a second highest single cell temperature corresponding to the charging and discharging section, and determine the second temperature mean value, the second maximum highest single cell temperature, the temperature mean value difference and the temperature maximum value difference as the statistical value calculation result of the second highest single cell temperature; and perform classification marking on the corresponding second highest single cell temperature according to multiple statistical value calculation results, to obtain temperature feature data corresponding to each of the multiple charging and discharging sections.

[0100] Optionally, the data processing module 302 is specifically configured to perform the following operations for the statistical value calculation result of each second highest single cell temperature: in a case that the statistical value calculation result is greater than a preset fault threshold, mark the second highest single cell temperature as abnormal, and determine abnormal state data as the temperature feature data; and in a case that the statistical value calculation result is less than or equal to the preset fault threshold, mark the second highest single cell temperature as normal, and determine normal state data as the temperature feature data.

[0101] Optionally, the data processing module 302 is specifically configured to obtain multiple sub-temperatures included in the second highest single cell temperature corresponding to the first charging and discharging section, determine the first temperature mean value according to values of the multiple sub-temperatures, and determine a maximum temperature in the multiple sub-temperatures as the first maximum highest single cell temperature.

[0102] Optionally, the data processing module 302 is specifically configured to acquire a previous charging and discharging section of the other charging and discharging section, and a third temperature average value and a third maximum highest single cell temperature corresponding to the previous charging and discharging section; determine the temperature average value difference according to the second temperature average value and the third temperature average value; and determine the temperature maximum value difference according to the second maximum highest single cell temperature and the third maximum highest single cell temperature.

[0103] Optionally, the data processing module 302 is specifically configured to perform extreme value deletion and / or missing value processing on the plurality of first highest single cell temperatures to obtain a plurality of third highest single cell temperatures; and determine the second highest single cell temperature corresponding to each of the plurality of charging and discharging sections from the plurality of third highest single cell temperatures.

[0104] Optionally, the temperature early warning module 303 is specifically configured to input the plurality of temperature feature data, the plurality of total currents and the plurality of residual capacities as input data into a light gradient boosting machine (LGBM) model to obtain a fault early warning result of the power battery output by the LGBM model.

[0105] As shown in Figure 4 The electronic device provided by the application includes a processor 410, a communications interface 420, a memory 430 and a communications bus 440, wherein the processor 410, the communications interface 420 and the memory 430 complete mutual communication through the communications bus 440. The processor 410 can call logical instructions in the memory 430 to execute a fault early warning method of a power battery, which includes: collecting a plurality of data information of the power battery at equal intervals within a preset time length, the power battery including a plurality of battery packs, each of the data information including a first highest single cell temperature corresponding to the plurality of battery packs, a total current and a residual capacity of the power battery, the plurality of data information corresponding to a plurality of collection periods one by one; acquiring a plurality of charging and discharging sections of the power battery within the preset time length, and determining a second highest single cell temperature corresponding to each charging and discharging section from a plurality of first highest single cell temperatures; classifying and marking a sub-temperature included in the second highest single cell temperature according to a statistical value calculation result of each second highest single cell temperature to obtain temperature feature data, the sub-temperature belonging to the plurality of first highest single cell temperatures; and determining a fault early warning result of the power battery according to a plurality of temperature feature data, a plurality of total currents and a plurality of residual capacities.

[0106] In addition, the logic instructions in the memory 430 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0107] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor, so that the computer can execute the power battery fault early warning method provided by the above method, the method comprises: collecting a plurality of data information of a power battery at equal intervals within a preset time period, the power battery comprises a plurality of battery packs, each data information comprises a first highest single temperature corresponding to the plurality of battery packs, a total current and a remaining capacity of the power battery, and the plurality of data information corresponds to a plurality of collection periods one by one; a plurality of charge and discharge sections of the power battery within the preset time period are obtained, and a second highest single temperature corresponding to each charge and discharge section is determined from a plurality of first highest single temperatures; according to the statistical value calculation result of each second highest single temperature, the sub-temperature included in the second highest single temperature is classified and marked to obtain temperature feature data, and the sub-temperature belongs to the plurality of first highest single temperatures; according to a plurality of temperature feature data, a plurality of total currents and a plurality of remaining capacities, a fault early warning result of the power battery is determined.

[0108] In yet another aspect, the present application also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, implements a method for early warning of a fault of a power battery, the method comprising: collecting a plurality of data information of the power battery at equal intervals within a preset time period, the power battery comprising a plurality of battery packs, each of the data information comprising a first highest single cell temperature corresponding to the plurality of battery packs, a total current of the power battery, and a remaining power, the plurality of data information corresponding to a plurality of collection periods one by one; obtaining a plurality of charge and discharge sections of the power battery within the preset time period, and determining a second highest single cell temperature corresponding to each charge and discharge section from the plurality of first highest single cell temperatures; classifying and marking sub-temperatures included in the second highest single cell temperature according to a calculation result of a statistical value of each of the second highest single cell temperatures, to obtain temperature feature data, the sub-temperatures belonging to the plurality of first highest single cell temperatures; and determining a fault early warning result of the power battery according to a plurality of temperature feature data, a plurality of total currents, and a plurality of remaining powers.

[0109] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0110] From the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and necessary general hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0111] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for failure early warning of a power battery, characterized in that, The method comprises the following steps: Collecting multiple first maximum single cell temperatures of the power battery at equal intervals within a preset time length, and obtaining multiple charging and discharging sections of the power battery within the preset time length, and total currents and residual capacities of the power battery corresponding to the multiple charging and discharging sections respectively, the multiple first maximum single cell temperatures corresponding to multiple collection periods one by one; wherein the first maximum single cell temperature is the maximum value of multiple single cell temperatures corresponding to multiple battery packs respectively; Determining second maximum single cell temperatures corresponding to the multiple charging and discharging sections from the multiple first maximum single cell temperatures; Traversing the multiple charging and discharging sections, for a current charging and discharging section, in a case that the current charging and discharging section is a first charging and discharging section, determining a first temperature mean value and a first maximum maximum single cell temperature according to the second maximum single cell temperature corresponding to the first charging and discharging section, and determining the first temperature mean value and the first maximum maximum single cell temperature as a statistical value calculation result of the second maximum single cell temperature; in a case that the current charging and discharging section is a charging and discharging section other than the first charging and discharging section, determining a second temperature mean value, a second maximum maximum single cell temperature, a temperature mean value difference and a temperature maximum value difference according to the second maximum single cell temperature corresponding to the charging and discharging section, and determining the second temperature mean value, the second maximum maximum single cell temperature, the temperature mean value difference and the temperature maximum value difference as the statistical value calculation result of the second maximum single cell temperature; According to multiple statistical value calculation results, the corresponding second maximum single cell temperature is classified and labeled to obtain temperature feature data corresponding to the multiple charging and discharging sections respectively; According to multiple temperature feature data, multiple total currents and multiple residual capacities, a fault warning result of the power battery is determined.

2. The method of claim 1, wherein, The statistical value calculation result of each second maximum single cell temperature is classified and labeled to obtain temperature feature data corresponding to the multiple charging and discharging sections respectively, comprising: For the statistical value calculation result of each second maximum single cell temperature, the following operations are performed: In a case that the statistical value calculation result is greater than a preset fault threshold, the second maximum single cell temperature is labeled as abnormal, and abnormal state data is determined as the temperature feature data; In a case that the statistical value calculation result is less than or equal to the preset fault threshold, the second maximum single cell temperature is labeled as normal, and normal state data is determined as the temperature feature data.

3. The method of claim 2, wherein, The statistical value calculation result of each second maximum single cell temperature is classified and labeled to obtain temperature feature data corresponding to the multiple charging and discharging sections respectively, comprising: Obtaining multiple sub-temperatures included in the second maximum single cell temperature corresponding to the first charging and discharging section; According to the values of the multiple sub-temperatures, the first temperature mean value is determined; The maximum temperature in the multiple sub-temperatures is determined as the first maximum maximum single cell temperature.

4. The method of claim 2, wherein, According to the second maximum single cell temperature corresponding to the charging and discharging section, the temperature mean value difference and the temperature maximum value difference are determined, comprising: Obtaining a previous charging and discharging section of the charging and discharging section, and a third temperature mean value and a third maximum maximum single cell temperature corresponding to the previous charging and discharging section; determine the temperature mean difference according to the second temperature mean and the third temperature mean; determine the temperature maximum difference according to the second maximum highest monomer temperature and the third maximum highest monomer temperature.

5. The method according to any of claims 1-2, characterized by, The determining, from the plurality of first highest monomer temperatures, of the second highest monomer temperature corresponding to each of the plurality of charging and discharging sections comprises: The determining, from the plurality of first highest monomer temperatures, of the second highest monomer temperature corresponding to each of the plurality of charging and discharging sections comprises:

6. The method according to any one of claims 1-2, characterized in that, The determining, from the plurality of first highest monomer temperatures, of the second highest monomer temperature corresponding to each of the plurality of charging and discharging sections comprises: The determining, from the plurality of first highest monomer temperatures, of the second highest monomer temperature corresponding to each of the plurality of charging and discharging sections comprises:

7. A failure early warning device for a power battery, characterized in that, The determining, from the plurality of first highest monomer temperatures, of the second highest monomer temperature corresponding to each of the plurality of charging and discharging sections comprises: The determining, from the plurality of first highest monomer temperatures, of the second highest monomer temperature corresponding to each of the plurality of charging and discharging sections comprises: The determining, from the plurality of first highest monomer temperatures, of the second highest monomer temperature corresponding to each of the plurality of charging and discharging sections comprises: The data acquisition module is configured to: acquire, at equal intervals within a preset time length, a plurality of first highest monomer temperatures of a power battery, and acquire a plurality of charging and discharging sections of the power battery within the preset time length, and a total current and a remaining electric quantity of the power battery corresponding to each of the plurality of charging and discharging sections, the plurality of first highest monomer temperatures corresponding to a plurality of acquisition periods one by one; wherein the first highest monomer temperature is a maximum value in a plurality of monomer temperatures corresponding to each of a plurality of battery packs.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The data processing module is configured to: determine, from the plurality of first highest monomer temperatures, a second highest monomer temperature corresponding to each of the plurality of charging and discharging sections; traverse the plurality of charging and discharging sections, and for a current charging and discharging section, in a case where the current charging and discharging section is a first charging and discharging section, determine a first temperature mean and a first maximum highest monomer temperature according to the second highest monomer temperature corresponding to the first charging and discharging section, and determine the first temperature mean and the first maximum highest monomer temperature as a statistical value calculation result of the second highest monomer temperature; in a case where the current charging and discharging section is a charging and discharging section other than the first charging and discharging section, determine a second temperature mean, a second maximum highest monomer temperature, a temperature mean difference and a temperature maximum difference according to the second highest monomer temperature corresponding to the charging and discharging section, and determine the second temperature mean, the second maximum highest monomer temperature, the temperature mean difference and the temperature maximum difference as the statistical value calculation result of the second highest monomer temperature; and according to a plurality of statistical value calculation results, classify and label corresponding second highest monomer temperatures to obtain temperature characteristic data corresponding to each of the plurality of charging and discharging sections. 9.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The temperature warning module is configured to determine a fault warning result of the power battery according to a plurality of temperature characteristic data, a plurality of total currents and a plurality of remaining electric quantities. The processor implements the power battery fault warning method according to any one of claims 1 to 6 when executing the program. The computer program implements the power battery fault warning method according to any one of claims 1 to 6 when executed by the processor. The computer program implements the power battery fault warning method according to any one of claims 1 to 6 when executed by the processor.

Images