Electric vehicle battery abnormity detection method and device based on vehicle model data

Abstract

The invention provides an electric vehicle battery abnormity detection method and device based on vehicle model data, and the method comprises the steps: collecting the battery data of all vehicles from a battery management system of a target number of vehicles in a target vehicle model based on a target period, and carrying out the aggregation of the battery data of all vehicles, the method comprises the following steps: constructing a battery data joint distribution model of all vehicles by adopting a preset statistical method, analyzing the battery data joint distribution model to obtain a target area with the highest data density, obtaining a charge state data estimation error range of a target vehicle type under normal work based on the target area, and aiming at each target vehicle, obtaining a charge state data estimation error range of the target vehicle type under normal work; the method comprises the following steps: comparing battery data of a target vehicle collected in real time with a charge state data estimation error range, detecting to obtain an abnormal target vehicle, obtaining an abnormal branch of the abnormal target vehicle, generating a corresponding early warning, determining branch information of the abnormal branch, and classifying and diagnosing the abnormity of the target vehicle based on the branch information. And determining a fault type and a diagnosis suggestion.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of batteries, in particular to a battery abnormality detection method and device for electric vehicles based on vehicle model data. BACKGROUND

[0002] The estimation of the state of health (SOH) of a battery aims to quantify the performance degradation of the battery due to aging, and the goal of the state of charge (SOC) estimation is to obtain the remaining power in real time and accurately. Therefore, the estimation of the state of health and the state of charge of the battery of an electric vehicle is beneficial to a more comprehensive understanding of the state of the battery.

[0003] At present, the estimation of the state of health and the state of charge of the battery of an electric vehicle mainly relies on the data collected by a single vehicle, such as voltage, current, temperature and internal resistance, etc. These methods are usually based on an equivalent circuit model or an electrochemical model of the battery, and combine Kalman filtering, ampere-hour integration and other algorithms to estimate the state of charge.

[0004] However, due to the differences in individual batteries, sensor errors, different use environments and driving habits, the accuracy of the state of charge estimation and the battery health diagnosis based solely on the data of a single vehicle is generally poor, and it is difficult to find latent complex abnormal conditions. In addition, diagnosis is usually performed when a single vehicle has obvious fault codes or a sharp performance decline, which makes it difficult to achieve early warning and maintenance. SUMMARY

[0005] Therefore, the purpose of the present application is to provide a battery abnormality detection method and device for electric vehicles based on vehicle model data. By constructing a battery data joint distribution model, the distribution of the battery data of the entire fleet can be intuitively displayed, and the health status of the entire vehicle model battery can be macroscopically and intuitively reflected. By analyzing the battery data joint distribution model, the error range of the state of charge estimation under normal working conditions of the vehicle model can be accurately determined, the accuracy of the state of charge estimation and the battery health diagnosis is improved, abnormal branches deviating from the mainstream data trend can be automatically identified, and accurate early detection and diagnosis of abnormal vehicles can be achieved, thereby achieving high-precision and early warning of abnormal vehicles.

[0006] In a first aspect, an embodiment of the present application provides a battery abnormality detection method for electric vehicles based on vehicle model data, which comprises: collecting battery data of all vehicles from a battery management system of a target number of vehicles of a target vehicle model based on a target period; wherein the battery data comprises total battery voltage data and state of charge data; aggregating the battery data of all vehicles, and constructing a battery data joint distribution model of all vehicles using a preset statistical method; analyzing the battery data joint distribution model to obtain a target region with the highest data density, and obtaining a state of charge data estimation error range of the target vehicle under normal working based on the target region; For each target vehicle, the real-time collected battery data of the target vehicle is compared with the state of charge data estimation error range, an abnormal target vehicle is detected, an abnormal branch of the abnormal target vehicle is obtained, and a corresponding early warning is generated; The branch information of the abnormal branch is determined, and the abnormality of the target vehicle is classified and diagnosed based on the branch information to determine the corresponding fault type and diagnosis suggestion.

[0007] In one possible implementation, the state of charge data estimation error range of the target vehicle under normal working based on the target region includes: The center curve of the target region is determined based on polynomial regression or spline curve fitting; The corresponding confidence interval is determined based on the standard deviation of the center curve, and the distribution width on both sides of the target region is determined based on the confidence interval; wherein the distribution width is the state of charge data estimation error range of the target vehicle under normal working.

[0008] In one possible implementation, the real-time collected battery data of the target vehicle is compared with the state of charge data estimation error range, an abnormal target vehicle is detected, and an abnormal branch of the abnormal target vehicle is obtained, including: If a target data point in the battery data of the target vehicle continuously or multiple times falls outside the state of charge data estimation error range, it is determined that the target vehicle is abnormal; The historical data points of the target vehicle with abnormality are determined, and all target historical data points deviating from the state of charge data estimation error range in the historical data points are determined; If all target historical data points form an abnormal branch, the target vehicle is marked as abnormal.

[0009] In one possible implementation, the method further includes: If the overall downward shift of the curve of the battery data joint distribution model is greater than a preset downward shift threshold, it is determined that the battery of the target vehicle has capacity attenuation; If the dispersion degree of the curve of the battery data joint distribution model is greater than a preset dispersion degree threshold, it is determined that the state of charge data estimation of the target vehicle is unstable or the sensor drifts.

[0010] In one possible implementation, the classification and diagnosis of the abnormality of the target vehicle based on the branch information includes: aggregating battery data of all vehicles periodically; updating the battery data joint distribution model and the state of charge data estimation error range based on the aggregated battery data.

[0011] In a possible implementation, the method further includes: comparing a new data point of each vehicle with the state of charge data estimation error range in real time; determining deviation information of the new data point from the state of charge data estimation error range and recording the deviation information.

[0012] In a possible implementation, the method further includes: obtaining auxiliary parameters of a target number of vehicles of a target vehicle model; wherein the auxiliary parameters at least include current, temperature, mileage; generating a multi-dimensional battery data joint distribution model based on the auxiliary parameters and the battery data.

[0013] In a second aspect, the embodiments of the present application further provide an electric vehicle battery anomaly detection device based on vehicle model data, the device includes: a collection module configured to collect battery data of all vehicles from a battery management system of a target number of vehicles of a target vehicle model based on a target period; wherein the battery data includes total battery voltage data and state of charge data; a construction module configured to aggregate the battery data of all vehicles and construct a battery data joint distribution model of all vehicles by using a preset statistical method; a first acquisition module configured to analyze the battery data joint distribution model, obtain a target area with the highest data density, and obtain a state of charge data estimation error range of the target vehicle model under normal working based on the target area; a second acquisition module configured to compare, for each target vehicle, battery data of the target vehicle collected in real time with the state of charge data estimation error range, detect an abnormal target vehicle, obtain an abnormal branch of the abnormal target vehicle, and generate a corresponding early warning; a determination module configured to determine branch information of the abnormal branch, classify and diagnose the abnormality of the target vehicle based on the branch information, determine a corresponding fault type and diagnosis suggestion.

[0014] In a possible implementation, the first acquisition module is specifically configured to: determine a center curve of the target area based on polynomial regression or spline curve fitting; determine a confidence interval based on the standard deviation of the center curve, and determine a distribution width on both sides of the target region based on the confidence interval; wherein the distribution width is a state of charge data estimation error range of the target vehicle under normal working condition.

[0015] In a possible implementation, the second obtaining module is specifically configured to: determine that the target vehicle is abnormal in response to a target data point in the battery data of the target vehicle falling outside the state of charge data estimation error range continuously or multiple times; determine a historical data point of the target vehicle that is abnormal, and determine all target historical data points deviating from the state of charge data estimation error range in the historical data point; in response to all target historical data points forming an abnormal branch, mark the target vehicle as abnormal.

[0016] In a possible implementation, the apparatus further includes: a second determining module configured to determine that the battery of the target vehicle is in capacity attenuation in response to the curve of the battery data joint distribution model moving downward as a whole by more than a preset downward moving threshold value; a third determining module configured to determine that the state of charge data estimation of the target vehicle is unstable or the sensor is drifting in response to a curve dispersion degree of the battery data joint distribution model being greater than a preset dispersion degree threshold value.

[0017] In a possible implementation, the determining module is specifically configured to: aggregate the battery data of all vehicles periodically; update the battery data joint distribution model and the state of charge data estimation error range based on the aggregated battery data.

[0018] In a possible implementation, the apparatus further includes: a comparison module configured to compare a new data point of each vehicle with the state of charge data estimation error range in real time; a fourth determining module configured to determine deviation information of the new data point from the state of charge data estimation error range and record the deviation information.

[0019] In a possible implementation, the apparatus further includes: a third obtaining module configured to obtain auxiliary parameters of a target number of vehicles of a target vehicle model; wherein the auxiliary parameters at least include current, temperature, and mileage; a generating module configured to generate a multi-dimensional battery data joint distribution model based on the auxiliary parameters and the battery data.

[0020] In a third aspect, an electronic device is provided, and includes a processor, a storage medium, and a bus. The storage medium stores machine readable instructions executable by the processor. When the electronic device is running, the processor communicates with the storage medium through the bus. The processor executes the machine readable instructions to perform the steps of the method for detecting battery abnormalities of an electric vehicle based on vehicle model data according to any one of the first aspect.

[0021] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. When the computer program is run by a processor, the computer program performs the steps of the method for detecting battery abnormalities of an electric vehicle based on vehicle model data according to any one of the first aspect.

[0022] The method and device for detecting battery abnormalities of an electric vehicle based on vehicle model data provided by the embodiments of the present application collect battery data of all vehicles from battery management systems of a target number of vehicles of a target vehicle model based on a target period, aggregate the battery data of all vehicles, construct a joint distribution model of the battery data of all vehicles using a preset statistical method, analyze the joint distribution model of the battery data, obtain a target region with the highest data density, obtain a state of charge data estimation error range of the target vehicle model under normal working conditions based on the target region, compare, for each target vehicle, real-time collected battery data of the target vehicle with the state of charge data estimation error range, detect abnormal target vehicles, obtain abnormal branches of the abnormal target vehicles, generate corresponding early warnings, determine branch information of the abnormal branches, classify and diagnose the abnormalities of the target vehicles based on the branch information, and determine corresponding fault types and diagnosis suggestions. The embodiments of the present application can intuitively show the distribution of battery data of the entire vehicle fleet, macroscopically and intuitively reflect the battery health status of the entire vehicle model, accurately determine the state of charge estimation error range of the vehicle model under normal working conditions by analyzing the joint distribution model of the battery data, improve the accuracy of state of charge estimation and battery health diagnosis, automatically identify abnormal branches deviating from the mainstream data trend, and accurately early detect and diagnose abnormal vehicles, thereby achieving high-precision early warning of abnormal vehicles.

[0023] To make the above objectives, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are specifically described below with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as limiting the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0025] Figure 1 is a flow chart of the electric vehicle battery anomaly detection method based on vehicle type data provided by the embodiments of the present application; Figure 2 is a schematic diagram of determining an abnormal branch from a data point distribution; Figure 3 is a structural schematic diagram of the electric vehicle battery anomaly detection device based on vehicle type data provided by the embodiments of the present application; Figure 4 is a structural schematic diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. It should be understood that the drawings in the present application only play the purpose of illustration and description, and do not limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn according to the actual proportion. The flow chart shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow chart can not be implemented in sequence, and the steps without logical context relationship can be reversed in sequence or implemented simultaneously. In addition, one or more other operations can be added to the flow chart or removed from the flow chart under the guidance of the content of the present application by those skilled in the art.

[0027] In addition, the described embodiments are only some of the embodiments of the present application, not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0028] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0029] The estimation of the state of health (SOH) aims to quantify the performance degradation of the battery due to aging, and the estimation of the state of charge (SOC) aims to obtain the remaining power in real time and accurately. Therefore, the estimation of the state of health and the state of charge of the battery of the electric vehicle can help to comprehensively grasp the state of the battery.

[0030] At present, the estimation of the state of health and the state of charge of the battery of the electric vehicle mainly relies on the data collected by the vehicle, such as voltage, current, temperature and internal resistance. These methods are usually based on the equivalent circuit model or electrochemical model of the battery, and combine Kalman filtering, ampere-hour integration and other algorithms to estimate the state of charge.

[0031] However, due to the individual differences of the battery, sensor errors, different use environments and driving habits, the accuracy of the state of charge estimation and the battery health diagnosis based on the data of the single vehicle is generally poor, and it is difficult to find latent complex abnormal conditions. In addition, the diagnosis is usually performed when the single vehicle has obvious fault codes or performance sharply decreases, which makes it difficult to achieve early warning and maintenance.

[0032] To solve the problem, the application provides a battery abnormality detection method and device for electric vehicles based on vehicle data. By constructing a battery data joint distribution model, the distribution of the battery data of the entire vehicle fleet can be intuitively displayed, and the health status of the entire vehicle battery can be macroscopically and intuitively reflected. By analyzing the battery data joint distribution model, the estimation error range of the state of charge of the vehicle under normal working conditions can be accurately determined, the accuracy of the state of charge estimation and the battery health diagnosis is improved, the abnormal branches deviating from the mainstream data trend can be automatically identified, and the accurate early detection and diagnosis of the abnormal vehicle can be realized, thereby realizing the high-precision and early warning of the abnormal vehicle.

[0033] Figure 1 The flowchart of the battery abnormality detection method for electric vehicles based on vehicle data provided by the embodiment of the application is shown in FIG. 1. As shown in FIG. 1, the battery abnormality detection method for electric vehicles based on vehicle data provided by the embodiment of the application can specifically include the following steps. Figure 1 S101, collecting the battery data of all vehicles from the battery management systems of the target number of vehicles of the target vehicle model based on a target period.

[0034] S102, aggregating the battery data of all vehicles, and constructing a battery data joint distribution model of all vehicles by using a preset statistical method.

[0035] S103, analyzing the battery data joint distribution model to obtain a target region with the highest data density, and obtaining the estimation error range of the state of charge data of the target vehicle under normal working conditions based on the target region.

[0036] ​S104, for each target vehicle, compare the real-time collected battery data of the target vehicle with the state of charge data estimation error range, detect the abnormal target vehicle, obtain the abnormal branch of the abnormal target vehicle, and generate the corresponding early warning.

[0037] S105, determine the branch information of the abnormal branch, and classify and diagnose the abnormality of the target vehicle based on the branch information, determine the corresponding fault type and diagnosis suggestion.

[0038] In the above electric vehicle battery abnormality detection method based on vehicle type data, by constructing a battery data joint distribution model, the distribution of the entire fleet battery data can be intuitively displayed, and the health status of the entire vehicle type battery can be macroscopically and intuitively reflected. By analyzing the battery data joint distribution model, the state of charge estimation error range of the vehicle type under normal working conditions can be accurately determined, the accuracy of state of charge estimation and battery health degree diagnosis is improved, abnormal branches deviating from the mainstream data trend can be automatically identified, and accurate early detection and diagnosis of abnormal vehicles can be realized, thereby realizing high-precision and early warning of abnormal vehicles.

[0039] The above exemplary steps of the embodiments of the present application will be described below in conjunction with specific examples: S101, based on a target period, collect battery data of all vehicles from the battery management system of the target number of vehicles of the target vehicle type.

[0040] In the embodiments of the present application, the target period is the period of collecting battery data, the target vehicle type is the vehicle type to be detected for electric vehicle battery abnormality, the target number is the number of selected vehicles of the target vehicle type set, the battery data includes battery total voltage data (V pack) and state of charge data (State Of Charge, SOC), therefore, the battery data is voltage-SOC data; the battery management system is (Battery Management System, BMS), the battery data includes battery total voltage data and state of charge data, and the battery data of all vehicles is periodically collected from the battery management system of the target number of vehicles of the target vehicle type for subsequent processing.

[0041] S102, aggregate the battery data of all vehicles, and construct a battery data joint distribution model of all vehicles using a preset statistical method.

[0042] In the embodiments of the present application, the statistical method is a pre-set statistical method, for example, the statistical method at least includes a two-dimensional histogram, a kernel density estimation or a heat map generation; all battery data of the vehicles collected in step S101 are aggregated, and a joint distribution model of the battery data of all vehicles is constructed by using the pre-set statistical method for subsequent processing. The joint distribution model of the battery data can be a heat map, which is taken as an example for description in the present application, but does not constitute a limitation on the joint distribution model of the battery data.

[0043] In step S103, the joint distribution model of the battery data is analyzed to obtain a target region with the highest data density, and a state of charge data estimation error range of the target vehicle under normal working condition is obtained based on the target region.

[0044] In the embodiments of the present application, the state of charge data estimation error range represents a normal range of the state of charge of the target vehicle under normal working condition, which can be understood as a normal range model; the joint distribution model of the battery data in step S102 is analyzed to obtain a target region with the highest data density, and a state of charge data estimation error range of the target vehicle under normal working condition is obtained based on the target region for subsequent processing.

[0045] Optionally, when the state of charge data estimation error range of the target vehicle under normal working condition is obtained based on the target region, a center curve of the target region is determined based on polynomial regression or spline curve fitting; a corresponding confidence interval is determined based on the standard deviation of the center curve, and a distribution width on both sides of the target region is determined based on the confidence interval. The distribution width is the state of charge data estimation error range of the target vehicle under normal working condition.

[0046] It should be noted that by analyzing the target region (target region) with the highest data density in the joint distribution model of the battery data (two-dimensional heat map), the center curve (for example, using polynomial regression or spline curve fitting) and the distribution width (for example, determining a confidence interval based on the standard deviation) on both sides thereof are determined, and the distribution width is the SOC estimation error range (state of charge data estimation error range) of the target vehicle under normal working condition.

[0047] In step S104, for each target vehicle, the real-time collected battery data of the target vehicle is compared with the state of charge data estimation error range, an abnormal target vehicle is detected, an abnormal branch of the abnormal target vehicle is obtained, and a corresponding warning is generated.

[0048] In the embodiments of the present application, the abnormal branch is the trajectory of the abnormal data point, the real-time collected battery data of the target vehicle is compared with the state of charge data estimation error range, an abnormal target vehicle is detected, an abnormal branch of the abnormal target vehicle is obtained, and a corresponding warning is generated.

[0049] In some embodiments, in response to the target data point in the battery data of the target vehicle continuously or multiple times falling outside the state of charge data estimation error range, it is determined that the target vehicle has an anomaly; historical data points of the target vehicle with the anomaly are determined, all target historical data points deviating from the state of charge data estimation error range in the historical data points are determined; in response to all target historical data points forming an anomaly branch, the target vehicle is marked as abnormal.

[0050] It should be noted that for each vehicle, the voltage-SOC data points of the battery data collected in real time are compared with the state of charge data estimation error range. If a data point continuously or multiple times falls outside the normal range, it is determined that the vehicle has an anomaly. The system tracks the historical data points of the abnormal vehicle. If the data points deviating from the state of charge data estimation error range (target historical data points) form a stable and identifiable trajectory (i.e., an anomaly branch), the vehicle is marked as abnormal. For example, as shown in FIG. 2, the density distribution of data points is shown. The data points in the normal working state form a dense and certain width "data ridge", and the abnormal data deviates from the "data ridge" to form an "anomaly branch". The two regions in the middle represent the data distribution of most vehicles in the normal working state, and the other part of the region outside represents sparse data or abnormal points. The anomaly branch is clearly visible. Figure 2

[0051] S105, branch information of the anomaly branch is determined, and the anomaly of the target vehicle is classified and diagnosed based on the branch information, a corresponding fault type and a diagnosis suggestion are determined.

[0052] In the embodiments of the present application, the branch information at least includes the shape and position of the anomaly branch. The branch information of the anomaly branch of the target vehicle with the anomaly in the determination step S103 is determined, and the anomaly of the target vehicle is classified and diagnosed according to the branch information, a corresponding fault type and a diagnosis suggestion are determined.

[0053] Optionally, when classifying and diagnosing the anomaly of the target vehicle based on the branch information, the battery data of all vehicles is aggregated periodically; the battery data joint distribution model and the state of charge data estimation error range are updated based on the aggregated battery data.

[0054] ​It should be noted that the application can periodically (such as every day, every week) aggregate the historical battery data of all vehicles to generate the latest battery data joint distribution model and the state of charge data estimation error range. The battery abnormality detection method for electric vehicles based on vehicle type data provided by the embodiment of the application collects battery data of all vehicles from the battery management system of a target number of vehicles of a target vehicle type based on a target period, aggregates the battery data of all vehicles, and constructs a battery data joint distribution model of all vehicles using a preset statistical method. The battery data joint distribution model is analyzed to obtain a target area with the highest data density, and based on the target area, a state of charge data estimation error range of the target vehicle type under normal working conditions is obtained. For each target vehicle, the real-time collected battery data of the target vehicle is compared with the state of charge data estimation error range to detect abnormal target vehicles, obtain abnormal branches of the abnormal target vehicles, and generate corresponding early warnings. The branch information of the abnormal branch is determined, and the abnormality of the target vehicle is classified and diagnosed based on the branch information to determine the corresponding fault type and diagnosis suggestion. The battery abnormality detection method for electric vehicles based on vehicle type data can visually display the distribution of the battery data of the entire vehicle fleet, macroscopically and intuitively reflect the battery health status of the entire vehicle type, accurately determine the state of charge estimation error range of the vehicle type under normal working conditions by analyzing the battery data joint distribution model, improve the accuracy of state of charge estimation and battery health diagnosis, automatically identify abnormal branches that deviate from the mainstream data trend, and achieve accurate early detection and diagnosis of abnormal vehicles, thereby achieving high-precision early warning of abnormal vehicles.

[0055] Further, in response to the overall downward shift of the curve of the battery data joint distribution model being greater than a preset downward shift threshold, it is determined that the battery of the target vehicle has capacity attenuation; and in response to the dispersion degree of the curve of the battery data joint distribution model being greater than a preset dispersion threshold, it is determined that the state of charge data estimation is unstable or the sensor drifts.

[0056] It should be noted that in addition to identifying abnormal branches, the curve of the battery data joint distribution model itself can also be diagnosed. For example, when the overall curve of the battery data joint distribution model shifts downward and exceeds the downward shift threshold (for example, 3%-5%), it indicates that the battery has capacity attenuation, and a preliminary diagnosis suggestion that the battery has capacity attenuation can be given. When the dispersion degree of the curve of the battery data joint distribution model is too large, i.e., exceeds the dispersion threshold, it is determined that the state of charge data estimation is unstable or the sensor drifts, and a preliminary diagnosis suggestion that the state of charge data estimation is unstable or the sensor drifts can be given accordingly. The operation and maintenance personnel can prioritize detailed inspection of the vehicle based on this suggestion, thereby achieving preventive maintenance and avoiding more serious failures.

[0057] Further, the new data points of each vehicle are compared with the error range of the state of charge data in real time; deviation information of the new data points from the error range of the state of charge data is determined and recorded.

[0058] It should be noted that the new data points of each vehicle are compared with the error range of the state of charge data in real time, and the deviation is recorded to realize real-time monitoring.

[0059] Further, auxiliary parameters of a target number of vehicles of a target vehicle model are obtained; and a multi-dimensional battery data joint distribution model is generated based on the auxiliary parameters and the battery data. The auxiliary parameters at least include current, temperature, and mileage.

[0060] It should be noted that in addition to the battery total voltage data and the state of charge data (voltage and SOC data), current, temperature, mileage and other parameters can be added to form a multi-dimensional battery data joint distribution model to more accurately identify different types of abnormalities.

[0061] Optionally, clustering analysis, principal component analysis (PCA) and other machine learning algorithms can be used to automatically cluster and classify different "abnormal branches" to form an automatic fault library for subsequent diagnosis.

[0062] Figure 3 is a structural schematic diagram of an electric vehicle battery abnormality detection device based on vehicle model data provided by an embodiment of the present application; as Figure 3 The electric vehicle battery abnormality detection device 300 based on vehicle model data of the embodiment of the present application can specifically include: The collection module is configured to collect battery data of all vehicles from a battery management system of a target number of vehicles of a target vehicle model based on a target period; wherein the battery data includes battery total voltage data and state of charge data; The construction module is configured to aggregate the battery data of all vehicles and construct a battery data joint distribution model of all vehicles by using a preset statistical method; The first acquisition module is configured to analyze the battery data joint distribution model to obtain a target region with the highest data density, and obtain an error range of the state of charge data estimation of the target vehicle model under normal working conditions based on the target region; The second acquisition module is configured to compare the real-time collected battery data of the target vehicle with the error range of the state of charge data estimation for each target vehicle, detect abnormal target vehicles, obtain abnormal branches of the abnormal target vehicles, and generate corresponding early warnings; The determination module is configured to determine branch information of the abnormal branches, classify and diagnose the abnormalities of the target vehicle based on the branch information, and determine corresponding fault types and diagnosis suggestions.

[0063] In a possible implementation, the first obtaining module is specifically configured to: determine a center curve of the target region based on polynomial regression or spline curve fitting; determine a corresponding confidence interval based on a standard deviation of the center curve, and determine a distribution width on both sides of the target region based on the confidence interval; wherein the distribution width is an error range of the state of charge data estimation of the target vehicle under normal working conditions.

[0064] In a possible implementation, the second obtaining module is specifically configured to: in response to the target data point in the battery data of the target vehicle continuously or multiple times falling outside the error range of the state of charge data estimation, determine that the target vehicle is abnormal; determine historical data points of the target vehicle that are abnormal, and determine all target historical data points deviating from the error range of the state of charge data estimation in the historical data points; in response to all target historical data points forming an abnormal branch, mark the target vehicle as abnormal.

[0065] In a possible implementation, the device further includes: a second determining module configured to determine that the battery of the target vehicle has capacity attenuation in response to the curve of the battery data joint distribution model moving downward as a whole by more than a preset downward moving threshold value; a third determining module configured to determine that the state of charge data estimation of the target vehicle is unstable or the sensor drifts in response to a curve dispersion degree of the battery data joint distribution model being greater than a preset dispersion degree threshold value.

[0066] In a possible implementation, the determining module is specifically configured to: aggregate the battery data of all vehicles periodically; update the battery data joint distribution model and the error range of the state of charge data estimation based on the aggregated battery data.

[0067] In a possible implementation, the device further includes: a comparison module configured to compare, in real time, a new data point of each vehicle with the error range of the state of charge data estimation; a fourth determining module configured to determine deviation information of the new data point from the error range of the state of charge data estimation, and record the deviation information.

[0068] In a possible implementation, the device further includes: a third obtaining module configured to obtain auxiliary parameters of a target number of vehicles of a target vehicle model; wherein the auxiliary parameters at least include current, temperature, and mileage; a generating module configured to generate a multi-dimensional battery data joint distribution model based on the auxiliary parameters and the battery data.

[0069] The embodiment of the application provides a battery abnormality detection device for electric vehicles based on vehicle type data, battery data of all vehicles is collected from battery management systems of target number of vehicles of a target vehicle type based on a target period, the battery data of all vehicles is aggregated, a joint distribution model of the battery data of all vehicles is constructed by adopting a preset statistical method, the joint distribution model of the battery data is analyzed, a target area with the highest data density is obtained, a state of charge data estimation error range of the target vehicle type under normal working conditions is obtained based on the target area, for each target vehicle, the battery data of the target vehicle collected in real time is compared with the state of charge data estimation error range, an abnormal target vehicle is detected, an abnormal branch of the abnormal target vehicle is obtained, and corresponding early warning is generated, branch information of the abnormal branch is determined, and the abnormality of the target vehicle is classified and diagnosed based on the branch information, and corresponding fault type and diagnosis suggestion are determined. The battery abnormality detection device for electric vehicles based on vehicle type data can intuitively show the distribution of the battery data of the entire vehicle fleet, macroscopically and intuitively reflect the battery health status of the entire vehicle type, accurately determine the state of charge estimation error range of the vehicle type under normal working conditions by analyzing the joint distribution model of the battery data, improve the accuracy of the state of charge estimation and the battery health degree diagnosis, automatically identify the abnormal branch deviating from the mainstream data trend, and realize accurate early detection and diagnosis of the abnormal vehicle, so that high-precision early warning of the abnormal vehicle is realized.

[0070] As shown in Figure 4 The embodiment of the application provides an electronic device 400, which comprises a processor 401, a memory 402 and a bus, the memory 402 stores machine readable instructions executable by the processor 401, when the electronic device is running, the processor 401 and the memory 402 communicate through the bus, and the processor 401 executes the machine readable instructions to execute the steps of the battery abnormality detection method for electric vehicles based on vehicle type data.

[0071] Specifically, the memory 402 and the processor 401 can be general memory and processor, which are not limited here, when the processor 401 runs the computer program stored in the memory 402, the battery abnormality detection method for electric vehicles based on vehicle type data can be executed.

[0072] Corresponding to the battery abnormality detection method for electric vehicles based on vehicle type data, the embodiment of the application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed by the processor to execute the steps of the battery abnormality detection method for electric vehicles based on vehicle type data.

[0073] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the system and the device described above can refer to the corresponding process in the method embodiment, and will not be repeated in the present application. In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented by other means. The above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and the actual implementation can have another division manner, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed mutual elements can be indirect coupling or communication connection through some communication interface, device or module, which can be electrical, mechanical or other forms.

[0074] The modules described as separate components can or can not be physically separated, and the components shown as modules can or can not be physical units, i.e. can be located in one place or distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0075] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0076] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the deployment method described in each embodiment of the present application. The foregoing storage medium includes: U disk, mobile hard disk, ROM, RAM, magnetic disk or optical disk and various program code storage media.

[0077] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for detecting battery anomalies in electric vehicles based on vehicle model data, characterized in that, The method includes: Based on a target period, battery data of all vehicles is collected from the battery management system of a target number of vehicles in a target model; wherein, the battery data includes total battery voltage data and state of charge data; The battery data of all vehicles are aggregated, and a joint distribution model of the battery data of all vehicles is constructed using a pre-defined statistical method. The joint distribution model of the battery data is analyzed to obtain the target area with the highest data density, and the state of charge data estimation error range of the target vehicle under normal operation is obtained based on the target area. For each target vehicle, the real-time collected battery data of the target vehicle is compared with the estimated error range of the state of charge data to detect abnormal target vehicles, obtain the abnormal branches of the abnormal target vehicles, and generate corresponding warnings. The branch information of the abnormal branch is determined, and the abnormality of the target vehicle is classified and diagnosed based on the branch information to determine the corresponding fault type and diagnostic suggestions.

2. The method according to claim 1, characterized in that, The estimation error range of the state of charge data of the target vehicle under normal operation obtained based on the target region includes: The center curve of the target region is determined by using polynomial regression or spline curve fitting. The confidence interval is determined based on the standard deviation of the center curve, and the distribution width on both sides of the target region is determined based on the confidence interval; wherein, the distribution width is the estimation error range of the state of charge data of the target vehicle under normal operation.

3. The method according to claim 2, characterized in that, The step of comparing the real-time collected battery data of the target vehicle with the estimated error range of the state of charge data to detect abnormal target vehicles, and obtaining the abnormal branches of the abnormal target vehicles, includes: If the target data point in the battery data of the target vehicle falls outside the state of charge data estimation error range continuously or multiple times, then the target vehicle is determined to be abnormal. Determine the historical data points of the target vehicle with anomalies, and identify all target historical data points that deviate from the estimation error range of the state of charge data; If an abnormal branch is formed in response to all target historical data points, the target vehicle is marked as abnormal.

4. The method according to claim 1, characterized in that, The method further includes: If the overall downward shift of the curve of the joint distribution model of battery data is greater than a preset downward shift threshold, it is determined that the battery of the target vehicle has experienced capacity decay. If the dispersion of the curve of the joint distribution model of battery data is greater than a preset dispersion threshold, it is determined that the state of charge data estimation of the target vehicle is unstable or the sensor is drifting.

5. The method according to claim 1, characterized in that, The classification and diagnosis of anomalies in the target vehicle based on the branch information includes: Periodically aggregate the battery data of all the vehicles mentioned; The error range of the joint distribution model of the battery data and the state of charge data is updated based on the aggregated battery data.

6. The method according to claim 1, characterized in that, The method further includes: Real-time comparison of new data points for each vehicle with the state of charge data to estimate the error range; The deviation information between the new data point and the estimated error range of the state of charge data is determined and recorded.

7. The method according to claim 1, characterized in that, The method further includes: Obtain auxiliary parameters for a target number of vehicles within the target vehicle model; wherein, the auxiliary parameters include at least current, temperature, and mileage; A multi-dimensional joint distribution model of battery data is generated based on the auxiliary parameters and the battery data.

8. An electric vehicle battery anomaly detection device based on vehicle model data, characterized in that, The device includes: The data acquisition module is used to acquire battery data of all vehicles from the battery management system of a target number of vehicles in a target model based on a target cycle; wherein, the battery data includes total battery voltage data and state of charge data; The module is used to aggregate battery data from all vehicles and construct a joint distribution model of battery data from all vehicles using a pre-defined statistical method. The first acquisition module is used to analyze the joint distribution model of the battery data, obtain the target area with the highest data density, and obtain the estimated error range of the state of charge data of the target vehicle under normal operation based on the target area. The second acquisition module is used to compare the real-time collected battery data of the target vehicle with the estimated error range of the state of charge data for each target vehicle, detect abnormal target vehicles, acquire the abnormal branches of the abnormal target vehicles, and generate corresponding warnings. The determination module is used to determine the branch information of the abnormal branch, and classify and diagnose the abnormality of the target vehicle based on the branch information, and determine the corresponding fault type and diagnostic suggestions.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the electric vehicle battery anomaly detection method based on vehicle model data as described in any one of claims 1 to 7 are performed.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the electric vehicle battery anomaly detection method based on vehicle model data as described in any one of claims 1 to 7.

Images