Abnormality detection method, product and terminal

Abstract

The invention provides an anomaly detection method, a product and a terminal, and belongs to the technical field of anomaly detection. The method comprises the steps that according to output value samples corresponding to voltage detection devices of battery packs, adjustment kurtosis corresponding to the output value samples is determined, and the adjustment kurtosis is used for representing the abnormal degree of any output value sample; and determining an abnormal voltage detection device according to the adjustment kurtosis corresponding to each output value sample. The invention aims to accurately detect whether the voltage detection device is abnormal or not.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of anomaly detection, and specifically, to an anomaly detection method, product, and terminal. Background Art

[0002] With the rapid development of the new energy vehicle industry, power batteries, as the core power components of new energy vehicles, have also attracted much attention. During the use of power batteries, the voltage of the battery pack inside the power battery may be low, which may lead to problems such as abnormal battery capacity and vehicle breakdown. In severe cases, there is even a risk of internal short circuit of the battery. Therefore, in order to protect the safety of people and property, a voltage detection device is currently installed on the vehicle to detect the voltage of the power battery pack. However, if the voltage detection device is abnormal or fails, the ability to detect the voltage of the battery pack is lost. Therefore, a method that can accurately detect whether the voltage detection device is abnormal is urgently needed. Summary of the Invention

[0003] The embodiments of the present application provide an abnormality detection method, product, and terminal, which are intended to accurately detect whether a voltage detection device of a battery pack is abnormal.

[0004] In a first aspect, an embodiment of the present application provides an anomaly detection method, the method comprising: determining, based on the output value samples corresponding to the voltage detection devices of the respective battery packs, the adjusted kurtosis corresponding to the respective output value samples, wherein the adjusted kurtosis is used to characterize the degree of abnormality of any output value sample; An abnormal voltage detection device is determined according to the adjusted kurtosis corresponding to each output value sample.

[0005] Optionally, the method further includes: Pre-process the acquired monitoring data within the target time period to determine the target data; According to the target data, output value samples corresponding to each voltage detection device are determined.

[0006] Optionally, preprocessing the acquired monitoring data within the target time period to determine target data includes: Preprocessing the monitoring data within the target time period to determine the valid data; Among the valid data, data in which the high-voltage system is in a normal state is selected as the target data.

[0007] Optionally, preprocess the monitoring data within the target time period, including: Desensitize the monitoring data within the target time period.

[0008] Optionally, preprocessing the monitoring data within the target time period further includes: Filter the fault data in the monitoring data within the target time period and sort them by time.

[0009] Optionally, determining output value samples corresponding to each voltage detection device according to the target data includes: In the target data, the output values ​​corresponding to the respective voltage detection devices that are greater than a first threshold value and less than a second threshold value are used as the output value samples corresponding to the respective voltage detection devices.

[0010] Optionally, the output value is a voltage evaluation value determined after monitoring the voltage of each battery pack.

[0011] Optionally, determining the adjusted kurtosis corresponding to each output value sample according to the output value sample corresponding to the voltage detection device of each battery pack includes: Determining the weight corresponding to each output value in each output value sample; The adjusted kurtosis corresponding to each output value sample is determined according to the weight corresponding to each output value in each output value sample.

[0012] Optionally, respectively determining a weight corresponding to each output value in each output value sample includes: For any output value sample, the weight corresponding to each output value in the output value sample is determined according to the sample standard deviation, sample mean and sample variance of the output value sample.

[0013] Optionally, the calculation formula for the weights corresponding to each output value in any output value sample is:

[0014] in, is the first of the output value samples i output values The corresponding weight; is the sample standard deviation of the output value sample; is the sample mean of the output value samples; is the sample variance of the output value sample; A natural constant An exponential function with base .

[0015] Optionally, determining the adjusted kurtosis corresponding to each output value sample according to the weight corresponding to each output value in each output value sample includes: For any output value sample, determine the expectation of the fourth power of the sample weighted center distance of the output value sample according to the weights corresponding to each output value in the output value sample; The adjusted kurtosis of the output value sample is determined according to an expectation of the fourth power of the sample weighted center distance of the output value sample and the square of the sample variance of the output value sample.

[0016] Optionally, the calculation formula for the adjusted kurtosis of any output value sample is:

[0017] in, is the adjusted kurtosis; is the expectation of the fourth power of the sample weighted center distance of the output value sample; is the sample variance of the output value sample the square of n is the number of output values ​​in the output value sample; is the sample mean of the output value samples; is the first of the output value samples i output values; is the first of the output value samples i output values The corresponding weight.

[0018] Optionally, determining the abnormal voltage detection device according to the adjusted kurtosis corresponding to each output value sample includes: determining whether each output value sample satisfies an abnormality condition according to the adjusted kurtosis corresponding to each output value sample; When the abnormal condition is satisfied, the abnormal voltage detection device is determined.

[0019] Optionally, the method further includes: When the abnormal condition is not satisfied, it is determined that no abnormal voltage detecting device exists among the respective voltage detecting devices.

[0020] Optionally, determining whether each output value sample satisfies an abnormality condition according to an adjusted kurtosis corresponding to each output value sample includes: determining, based on the adjusted kurtosis of each output value sample, a standard deviation ratio of a first standard deviation of all adjusted kurtosis to a second standard deviation of all non-maximum adjusted kurtosis; When the standard deviation ratio is greater than a standard deviation ratio threshold, it is determined that each of the output value samples meets an abnormal condition.

[0021] Optionally, determining whether each output value sample satisfies an abnormality condition according to an adjusted kurtosis corresponding to each output value sample includes: determining, based on the adjusted kurtosis of each output value sample, a standard deviation ratio of a first standard deviation of all adjusted kurtosis to a second standard deviation of all non-maximum adjusted kurtosis; Determine whether each output value sample meets an abnormal condition based on the standard deviation ratio and each output value sample.

[0022] Optionally, determining whether each output value sample meets an abnormal condition according to the standard deviation ratio and each output value sample includes: When the standard deviation ratio is greater than the standard deviation ratio threshold, and among the output value samples, the square of the sample variance of the output value sample corresponding to the maximum adjusted kurtosis is less than the variance square threshold, it is determined that the output value samples meet the abnormal condition.

[0023] Optionally, determining whether each output value sample meets an abnormal condition according to the standard deviation ratio and each output value sample includes: When the standard deviation ratio is greater than a standard deviation ratio threshold, and the minimum sample capacity of each output value sample is greater than a sample capacity threshold, it is determined that each output value sample meets an abnormal condition.

[0024] Optionally, determining whether each output value sample meets an abnormal condition according to the standard deviation ratio and each output value sample includes: When the standard deviation ratio is greater than the standard deviation ratio threshold, the square of the sample variance of the output value sample corresponding to the maximum adjusted kurtosis in each output value sample is less than the variance square threshold, and the minimum sample capacity in each output value sample is greater than the sample capacity threshold, it is determined that each output value sample meets the abnormal condition.

[0025] Optionally, when the abnormal condition is met, determining the abnormal voltage detection device includes: The voltage detection device corresponding to the output value sample with the largest adjusted kurtosis is used as the abnormal voltage detection device, and the number of the battery group corresponding to the abnormal voltage detection device is output.

[0026] In a second aspect, an embodiment of the present application provides an electronic device comprising: at least one processor, and a memory, wherein the memory stores a computer program that can be run on the processor, wherein when the processor executes the computer program, the anomaly detection method described in the first aspect of the embodiment is executed.

[0027] In a third aspect, an embodiment of the present application provides a non-volatile readable storage medium, wherein the non-volatile readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the anomaly detection method described in the first aspect of the embodiment is executed.

[0028] In a fourth aspect, an embodiment of the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the anomaly detection method described in the first aspect of the embodiment.

[0029] In a fifth aspect, an embodiment of the present application provides a terminal, which is used to execute the anomaly detection method described in the first aspect of the embodiment.

[0030] Optionally, the terminal includes a cloud terminal and a vehicle.

[0031] Beneficial effects: This method determines the adjusted kurtosis corresponding to each output value sample according to the output value samples corresponding to the voltage detection device of each battery pack, and then determines the abnormal voltage detection device according to the adjusted kurtosis corresponding to each output value sample.

[0032] By determining the adjusted kurtosis of the output value samples of each voltage detection device, quantifying the degree of abnormality of the output value samples of a voltage detection device based on the adjusted kurtosis, and finally determining the abnormal voltage detection device based on the adjusted kurtosis abnormality detection algorithm, it is possible to accurately detect whether the voltage detection device used by the vehicle to detect the battery pack voltage is abnormal, thereby improving the accuracy of abnormality detection of the battery pack voltage detection device. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0034] Figure 1 This is a flowchart of the steps of the anomaly detection method proposed in one embodiment of the present application; Figure 2 is a schematic diagram of output value samples of an abnormal voltage detection device proposed in one embodiment of the present application; Figure 3 2 is a schematic diagram of adjusting the kurtosis corresponding to the abnormal voltage detection device provided in one embodiment of the present application; Figure 4 This is a flowchart of an execution of an anomaly detection method proposed in one embodiment of the present application; Figure 5 This is a functional module diagram of an anomaly detection device proposed in one embodiment of the present application; Figure 6 is a schematic diagram of an electronic device provided in one embodiment of the present application; Figure 7 is a schematic diagram of a non-volatile readable storage medium proposed in an embodiment of the present application; Figure 8 It is a schematic diagram of a computer program product proposed in one embodiment of the present application. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0036] With the rapid development of the new energy vehicle industry, power batteries, as the core power components of new energy vehicles, have gradually attracted the attention of consumers. Although power batteries have many advantages such as large capacity, high energy density, and long service life, due to their relatively complex internal structure and production process, the voltage of the battery module or battery pack (BC) in the power battery may be low during use, resulting in abnormal battery capacity and vehicle breakdown. In severe cases, there is even the risk of internal battery short circuit. Therefore, in order to protect the safety of people and property, a method to determine whether the battery pack voltage in the power battery is abnormal is essential.

[0037] Methods for detecting abnormal battery pack voltage in power batteries can be divided into cloud-based detection solutions based on data uploaded by the vehicle to a cloud server, and vehicle-side detection solutions based on a voltage detection device equipped on the vehicle to detect the voltage of the battery pack in the power battery mold. However, if the voltage detection device is abnormal or fails, the ability to detect the voltage of the battery pack is lost. Therefore, a method that can accurately detect whether the voltage detection device is abnormal is urgently needed.

[0038] Therefore, the embodiment of the present application provides an abnormality detection method, which can accurately detect whether the voltage detection device of the battery pack is abnormal, thereby also improving the accuracy of detecting battery pack voltage abnormalities in the power battery.

[0039] Reference Figure 1 , shows a flowchart of the steps of an abnormality detection method in an embodiment of the present application, and the method may specifically include the following steps: S101: determining the adjusted kurtosis corresponding to each output value sample according to the output value samples corresponding to the voltage detection devices of each battery pack, wherein the adjusted kurtosis is used to characterize the abnormality degree of any output value sample.

[0040] In the actual implementation process, the vehicle's power battery includes multiple battery modules or battery packs. A corresponding voltage detection device, namely a BC voltage detection device, can be preset for each battery pack. Each battery pack is preset with a corresponding voltage detection device for detecting and evaluating the voltage output by the battery pack. The output of the voltage detection device is the output value.

[0041] Specifically, the output value of the voltage detection device can be a voltage value after monitoring and data processing the voltage of the battery pack, and the output value can also be a voltage evaluation value determined after monitoring the voltage of the battery pack. For example, according to the needs of actual applications, an evaluation strategy for the voltage output by each voltage detection device of the battery pack can be configured in the vehicle's BMS (Battery Management System). According to the evaluation strategy, it is determined how the voltage detection device monitors and evaluates the battery pack voltage and outputs the output value. The content and meaning of the output value can be selected according to the needs of actual applications, and this embodiment does not impose any restrictions.

[0042] This embodiment further provides a method for constructing output value samples of each voltage detection device. Specifically, the process of determining the output value samples corresponding to each voltage detection device may include the following steps: A1: Preprocess the acquired monitoring data within the target time period to determine the target data.

[0043] During the actual implementation process, the vehicle data of the vehicle will be uploaded to the cloud big data platform, and the monitoring data of each vehicle within the target time period can be obtained from the cloud big data platform. Other methods can also be used to obtain the monitoring data within the target time period, which is not limited in this embodiment.

[0044] The length of the target time period can be set according to actual application requirements. For example, monitoring data for at least 15 natural days can be obtained.

[0045] When pre-processing the acquired monitoring data, the monitoring data may be desensitized first. For example, when processing sensitive information such as vehicle user information, the scope of desensitization may be customized according to the needs of the actual application.

[0046] Furthermore, it is also possible to filter the fault data in the monitoring data within the target time period. For example, the fault data caused by vehicle-side software and hardware sampling failures, cloud communication failures and parsing errors are eliminated, and then the data is sorted by time. For data with the same data time, duplicate data is filtered out, and only one frame of multiple data with the same time is retained.

[0047] In actual application, the preprocessing process of monitoring data can be set according to the needs of actual application, and the monitoring data after preprocessing can be used as valid data.

[0048] Then, from the valid data, the data when the vehicle's high-voltage system is in a normal state is selected as the target data. By selecting the data when the high-voltage system is in a normal state as the target data, temporary abnormal values ​​that appear when the contactor switches between power on and power off can be avoided, and the temporary abnormal values ​​that appear can be avoided from affecting the accuracy of the abnormal judgment of the voltage detection device.

[0049] For example, it can be determined that the vehicle's high-voltage system is in a normal state based on the status of the positive and negative contactors in the vehicle's high-voltage system and the insulation resistance value. Taking the contactor status as an example, when the vehicle includes three contactors, based on the combination of the disconnection and attraction states of each contactor, such as 0 for disconnection and 1 for attraction, the state combination of the three contactors can be expressed as 000, 010..., and the state combination of the three contactors corresponding to the high-voltage system being in a normal state can be preset. Then, when the state combination of the three contactors does not represent that the high-voltage system is in a normal state, the frame data can be discarded.

[0050] A2: Determine the output value samples corresponding to each voltage detection device according to the target data.

[0051] Specifically, when the range of output values ​​of the voltage detection device of a battery pack is large, the range of output values ​​that contributes most to determining whether the voltage detection device is abnormal can be selected as the output value sample of the voltage detection device. For example, a first threshold and a second threshold are set, and the output value within the interval of the first threshold and the second threshold can be used to determine whether the voltage detection device is abnormal.

[0052] That is, in the target data, the output value corresponding to each voltage detection device that is greater than the first threshold and less than the second threshold is used as the output value sample corresponding to each voltage detection device, and the output value samples corresponding to all voltage detection devices constitute a sample set. ,in, For the j The output value samples of the voltage detection device.

[0053] By preprocessing the monitoring data, in addition to reducing the impact of noise data such as fault data on the abnormal detection of the voltage detection device, the accuracy of the abnormal detection of the voltage detection device is improved. It also greatly compresses the amount of data that needs to be processed during the abnormal detection process of the voltage detection device, and excludes abnormal data and fault data sent to the cloud when the voltage detection device is not working, effectively improving the detection efficiency and timeliness.

[0054] Reference Figure 2, a schematic diagram of an output value sample of the abnormal voltage detection device provided in an embodiment of the present application is shown. It is known that the voltage detection device of the No. 6 battery pack is an abnormal voltage detection device. According to observation results, when the voltage detection device of the No. 6 battery pack fails due to an abnormality, the distribution of the output value of the voltage detection device will be more concentrated.

[0055] Therefore, after determining the output value samples of each voltage detection device, the adjusted kurtosis corresponding to each output value sample can be determined respectively. The adjusted kurtosis can more accurately characterize the degree of central aggregation of the output value of any output value sample, and then the adjusted kurtosis can be used to characterize the degree of abnormality of an output value sample, thereby determining whether the voltage detection device corresponding to the output value sample is failed or abnormal.

[0056] Reference Figure 3 , shows a schematic diagram of the adjusted kurtosis corresponding to the abnormal voltage detection device provided in an embodiment of the present application. According to the observation results, when the voltage detection device of the No. 6 battery pack fails, the more concentrated the distribution of its output value is, the greater the adjusted kurtosis of the output value sample of the voltage detection device is.

[0057] The common mathematical definition of kurtosis is the expectation of the fourth power of the sample center distance and the square of the sample variance The ratio is calculated as follows:

[0058] Kurtosis can be used to measure the degree to which the data in a sample are concentrated in the center. The larger the kurtosis, the more concentrated the data in the sample are around the mean; the smaller the kurtosis, the less concentrated the data in the sample are around the mean. The kurtosis value of data that conforms to the standard normal distribution is 3.

[0059] However, traditional kurtosis has two defects: One is that when the data distribution in the sample is too concentrated, the kurtosis of the data set is smaller, because when the data distribution is too concentrated, the distance from the sample center to the fourth power is less than the expected value. The value is extremely small, and the square of the sample variance is is also extremely small, and the ratio of the two minimum values, that is, the kurtosis, will also be small.

[0060] The second is: According to the kurtosis formula, the kurtosis is affected by the square of the sample variance. and the expectation of the fourth power of the sample center distance Two factors affect the distance from the sample mean. When there is some data at the far end, although the amount of data is very small, The data are greatly affected, but the square of the sample variance It is basically unaffected, that is, the outliers far away from the sample mean dominate the change of kurtosis.

[0061] Therefore, this embodiment proposes a modified kurtosis method. The mathematical definition of modified kurtosis is the expected value of the fourth power of the sample weighted center distance. and the square of the sample variance ratio.

[0062] In a feasible implementation, the process of determining the adjusted kurtosis corresponding to each output value sample according to the output value sample corresponding to the voltage detection device of each battery pack includes the following steps: B1: respectively determining the weight corresponding to each output value in each output value sample.

[0063] Specifically, for any output value sample, the weight corresponding to each output value in the output value sample is determined according to the sample standard deviation, sample mean and sample variance of the output value sample.

[0064] For example, the calculation formula for the weight corresponding to each output value in any output value sample is:

[0065] in, is the first of the output value samples i output values The corresponding weight; is the sample standard deviation of the output value sample; is the sample mean of the output value samples; is the sample variance of the output value sample; A natural constant An exponential function with base .

[0066] B2: Determine the adjusted kurtosis corresponding to each output value sample according to the weight corresponding to each output value in each output value sample.

[0067] Specifically, for any output value sample, the expectation of the fourth power of the sample weighted center distance of the output value sample is determined based on the weights corresponding to each output value in the output value sample, and then the adjusted kurtosis of the output value sample is determined based on the expectation of the fourth power of the sample weighted center distance of the output value sample and the square of the sample variance of the output value sample.

[0068] For example, the calculation formula for the adjusted kurtosis of any output value sample is:

[0069] in, is the adjusted kurtosis; is the expectation of the fourth power of the sample weighted center distance of the output value sample; is the sample variance of the output value sample the square of n is the number of output values ​​in the output value sample; is the sample mean of the output value samples; is the first of the output value samples i output values; is the first of the output value samples i output values The corresponding weight.

[0070] Based on weight and adjusted kurtosis The calculation formula shows that the weight of the output value in the output value sample The output value sample has a normal distribution with a sample mean and variance of 1. The probability density function is calculated, and the output value closer to the sample mean is given a larger weight to the fourth power of the center distance, which can reduce the distance from the sample mean. The small amount of data in the distant tail adjusts the kurtosis The influence of can more accurately characterize the central aggregation degree of the output value in the output value sample, and then the abnormality detection of the voltage detection device can be more accurately performed based on the adjustment of the kurtosis.

[0071] S102: Determine an abnormal voltage detection device according to the adjusted kurtosis corresponding to each output value sample.

[0072] Specifically, after determining the adjusted kurtosis corresponding to each output value sample, it can be determined whether each output value sample meets the abnormal condition based on the adjusted kurtosis corresponding to each output value sample.

[0073] When the abnormal condition is met, it indicates that there is an abnormal voltage detection device among the voltage detection devices, and the abnormal voltage detection device can be further determined.

[0074] When the abnormal condition is not met, it indicates that there is no abnormal voltage detection device among the voltage detection devices, and the current abnormal detection can be ended.

[0075] In a feasible implementation manner, the process of determining whether each output value sample meets an abnormality condition according to the adjusted kurtosis corresponding to each output value sample includes: According to the adjusted kurtosis of each output value sample, a standard deviation ratio of a first standard deviation of all adjusted kurtosis to a second standard deviation of all non-maximum adjusted kurtosis is determined.

[0076] Specifically, first calculate the first standard deviation of all adjusted kurtosis , calculate the second standard deviation of all non-maximum adjusted kurtosis , that is, calculate the maximum adjusted kurtosis The second standard deviation of the remaining adjusted kurtosis .

[0077] Then, calculate the first standard deviation and the second standard deviation The standard deviation ratio R :

[0078] When the standard deviation ratio R When the value is greater than the standard deviation ratio threshold, it is determined that each output value sample meets the abnormal condition.

[0079] When the standard deviation ratio R When the value is less than or equal to the standard deviation ratio threshold, it is determined that each output value sample does not meet the abnormal condition.

[0080] Specifically, the standard deviation ratio can characterize the degree of outlier of the maximum adjusted kurtosis. If the standard deviation ratio is greater than the standard deviation ratio threshold, the maximum adjusted kurtosis is judged as an outlier. At this time, there is an abnormality in the voltage detection device corresponding to the maximum adjusted kurtosis. The standard deviation ratio threshold can be set according to the needs of actual application, and this embodiment does not impose any restrictions.

[0081] In another feasible implementation, determining whether each output value sample satisfies an abnormality condition according to the adjusted kurtosis corresponding to each output value sample may include: First, according to the adjusted kurtosis of each output value sample, the first standard deviation of all adjusted kurtosis is determined. and the second standard deviation of all non-maximum adjusted kurtosis The standard deviation ratio R :

[0082] Then, according to the standard deviation ratio and the respective output value samples, it is determined whether the respective output value samples meet an abnormal condition.

[0083] Furthermore, when determining whether each output value sample meets the abnormal condition based on the standard deviation ratio and the output value samples, the output value sample corresponding to the maximum adjusted kurtosis can be first determined among the adjusted kurtosis of the output value samples corresponding to all voltage detection devices.

[0084] For example, according to the output value samples corresponding to all voltage detection devices Adjusted kurtosis , determine the maximum adjusted kurtosis , and determine the maximum adjusted kurtosis Corresponding module output value .

[0085] Then, calculate the maximum adjusted kurtosis Corresponding module output value The square of the sample variance .

[0086] When the standard deviation ratio R When the output value sample is greater than the standard deviation ratio threshold and the maximum adjusted kurtosis The square of the sample variance When the value is less than the squared variance threshold, it is determined that each output value sample meets the abnormal condition.

[0087] That is, in addition to the standard deviation ratio R In addition to determining whether the maximum adjusted kurtosis is outlier based on the size relationship with the standard deviation ratio threshold, we can also further determine whether the maximum adjusted kurtosis is outlier based on the output value sample corresponding to the maximum adjusted kurtosis. The square of the sample variance , evaluate the output value sample corresponding to the maximum adjusted kurtosis The discrete degree of the output value sample can be reduced The adjusted kurtosis anomaly caused by the outlier element in The square of the sample variance If the value is smaller than the squared variance threshold, the credibility of the adjusted kurtosis is higher.

[0088] According to the standard deviation ratio R The size of the standard deviation ratio threshold and the output value sample corresponding to the maximum adjusted kurtosis The square of the sample variance The size of the squared deviation threshold can further improve the accuracy of abnormal detection of the voltage detection device.

[0089] In actual implementation, the sizes of the standard deviation ratio threshold and the variance square threshold can be set according to the needs of actual application, and this embodiment does not impose any restrictions.

[0090] Furthermore, when determining whether each output value sample satisfies an abnormal condition based on the standard deviation ratio and each output value sample, the standard deviation ratio may be determined. R Finally, the accuracy of abnormality detection of the voltage detection device can be improved according to the data capacity of each output value sample.

[0091] Specifically, the number of output values ​​in each output value sample is used as the sample capacity of each output value sample, and the minimum sample capacity is determined , For the i The number of output values ​​in a sample of output values.

[0092] When the standard deviation ratio R When the standard deviation ratio threshold is greater than the minimum sample size of each output value sample When the value is greater than the sample capacity threshold, it is determined that each output value sample meets the abnormal condition.

[0093] When the minimum sample capacity of the output value samples of each voltage detection device When it is also greater than the sample capacity threshold, it indicates that the abnormal judgment of the voltage detection device this time has a sample with a large enough capacity, which can ensure that the confidence and accuracy of the statistics are high enough, thereby further improving the accuracy of abnormal detection of the voltage detection device.

[0094] In actual implementation, the sizes of the standard deviation ratio threshold and the sample capacity threshold can be set according to the needs of actual application, and this embodiment does not impose any restrictions.

[0095] Furthermore, when determining whether each output value sample meets the abnormal condition based on the standard deviation ratio and each output value sample, the standard deviation ratio can also be used to determine whether each output value sample meets the abnormal condition. R , the output value sample corresponding to the maximum adjusted kurtosis The square of the sample variance and the minimum sample size to make a comprehensive judgment.

[0096] Specifically, when the standard deviation ratio R The output value sample that is greater than the standard deviation ratio threshold and corresponds to the maximum adjusted kurtosis among all output value samples The square of the sample variance Less than the squared variance threshold, and the minimum sample size in each output value sample When the value of each output value sample is greater than the sample capacity threshold, it is determined that each output value sample meets the abnormal condition, which can further improve the accuracy of abnormality detection of the voltage detection device.

[0097] In actual implementation, the sizes of the standard deviation ratio threshold, the variance square threshold, and the sample capacity threshold can be set according to the needs of actual application, and this embodiment does not impose any restrictions.

[0098] When the abnormal condition is met, the abnormal voltage detection device is determined to have the maximum adjusted kurtosis. Output value sample The corresponding voltage detection device serves as the abnormal voltage detection device and outputs the number of the battery pack corresponding to the abnormal voltage detection device.

[0099] Reference Figure 4, shows an execution flow chart of the anomaly detection method provided by an embodiment of the present application. In a feasible implementation, the method may include the following steps: S1: Obtain monitoring data within the target time period.

[0100] For example, monitoring data of vehicles within a target time period can be obtained on a cloud big data platform.

[0101] S2: Preprocess the monitoring data to obtain valid data.

[0102] The preprocessing process may include: desensitizing the monitoring data, filtering the fault data in the monitoring data, sorting the data by time, and deduplicating the duplicate data with the same time.

[0103] S3: Select data in which the high-voltage system is in a normal state from the valid data as target data.

[0104] For example, it can be determined that the high-voltage system is in a normal state based on the states of the positive and negative contactors and the insulation resistance values ​​in the high-voltage system of the vehicle.

[0105] S4: In the target data, the output values ​​of the respective voltage detection devices that are greater than the first threshold value and less than the second threshold value are taken as the output value samples of the respective voltage detection devices.

[0106] S5: Determine the adjusted kurtosis corresponding to each output value sample respectively.

[0107] Specifically, the formula for calculating the adjusted kurtosis corresponding to any output value sample is:

[0108]

[0109] in, is the adjusted kurtosis; is the expectation of the fourth power of the sample weighted center distance of the output value sample; is the sample variance of the output value sample The square of n is the number of output values ​​in the output value sample; is the sample mean of the output value samples; is the first of the output value samples i output values; is the first of the output value samples i output values The corresponding weight.

[0110] S6: Determine the output value sample corresponding to the maximum adjusted kurtosis.

[0111] S7: Calculate the standard deviation ratio of the first standard deviation of all adjusted kurtosis to the second standard deviation of all non-maximum adjusted kurtosis.

[0112] S8: Calculate the square of the sample variance of the output value sample corresponding to the maximum adjusted kurtosis.

[0113] S9: Determine the minimum sample size among all output value samples.

[0114] S10: Determine whether an abnormal condition is met.

[0115] For example, when the standard deviation ratio is greater than the standard deviation ratio threshold, the square of the sample variance of the output value sample with the maximum adjusted kurtosis is less than the variance square threshold, and the minimum sample capacity in each output value sample is greater than the sample capacity threshold, it is determined that each output value sample meets the abnormal condition, and S11 is executed when the abnormal condition is met, otherwise it ends.

[0116] S11: The voltage detection device corresponding to the output value sample with the maximum adjusted kurtosis is used as the abnormal voltage detection device, and the number of the battery group corresponding to the abnormal voltage detection device is output.

[0117] This method has at least the following beneficial effects: 1. After pre-processing the monitoring data of the vehicle within the target time period, the impact of fault data on the abnormal detection of the voltage detection device is reduced, and data when the vehicle's high-voltage system is in a normal state is selected as the data for abnormal detection, further reducing the impact of abnormal values ​​temporarily appearing when the contactor switches between power on and off on the abnormal detection of the voltage detection device. This not only improves the accuracy of the abnormal judgment of the voltage detection device, but also compresses the amount of data that needs to be processed during the abnormal detection process, which can improve the efficiency of the abnormal detection process.

[0118] 2. Based on the adjustment of kurtosis, the degree of central aggregation of the output value samples of each voltage detection device, that is, the degree of abnormality, is quantified, which significantly improves the accuracy of abnormality detection of the voltage detection device.

[0119] In actual verification, the accuracy of this method in detecting abnormalities in voltage detection devices is 93.75%.

[0120] In actual implementation, this method can be executed on the cloud to perform abnormality detection on the voltage detection device based on the strong computing power of the cloud, or the vehicle can execute this method to perform abnormality detection on its own voltage detection device, which is not limited in this embodiment.

[0121] Reference Figure 5 , shows a functional module diagram of an anomaly detection device provided in an embodiment of the present application, the device comprising: An adjusted kurtosis determination module 100 is configured to determine the adjusted kurtosis corresponding to each output value sample based on the output value samples corresponding to the voltage detection devices of each battery pack, wherein the adjusted kurtosis is used to characterize the abnormality of any output value sample; The abnormality detection module 200 is configured to determine an abnormal voltage detection device according to the adjusted kurtosis corresponding to each output value sample.

[0122] Optionally, the device further includes an output value sample determination module, configured to: A target data determination unit is used to pre-process the acquired monitoring data within the target time period to determine the target data; The output value sample determination unit is used to determine the output value samples corresponding to each voltage detection device according to the target data.

[0123] Optionally, the target data determination unit is configured to: Preprocessing the monitoring data within the target time period to determine the valid data; Among the valid data, data in which the high-voltage system is in a normal state is selected as the target data.

[0124] Optionally, the target data determination unit is configured to: Desensitize the monitoring data within the target time period.

[0125] Optionally, the target data determination unit is configured to: Filter the fault data in the monitoring data within the target time period and sort them by time.

[0126] Optionally, the output value sample determination unit is configured to: In the target data, the output values ​​corresponding to the respective voltage detection devices that are greater than a first threshold value and less than a second threshold value are used as the output value samples corresponding to the respective voltage detection devices.

[0127] Optionally, the output value is a voltage evaluation value determined after monitoring the voltage of each battery pack.

[0128] Optionally, the adjusted kurtosis determination module includes: A weight calculation unit, configured to determine the weight corresponding to each output value in each output value sample; The adjusted kurtosis calculation unit is used to determine the adjusted kurtosis corresponding to each output value sample according to the weight corresponding to each output value in each output value sample.

[0129] Optionally, the weight calculation unit is configured to: For any output value sample, the weight corresponding to each output value in the output value sample is determined according to the sample standard deviation, sample mean and sample variance of the output value sample.

[0130] Optionally, the calculation formula for the weights corresponding to each output value in any output value sample is:

[0131] in, is the first of the output value samples i output values The corresponding weight; is the sample standard deviation of the output value sample; is the sample mean of the output value samples; is the sample variance of the output value sample; A natural constant An exponential function with base .

[0132] Optionally, the adjusted kurtosis calculation unit is used to: For any output value sample, determine the expectation of the fourth power of the sample weighted center distance of the output value sample according to the weights corresponding to each output value in the output value sample; The adjusted kurtosis of the output value sample is determined according to an expectation of the fourth power of the sample weighted center distance of the output value sample and the square of the sample variance of the output value sample.

[0133] Optionally, the calculation formula for the adjusted kurtosis of any output value sample is:

[0134] in, is the adjusted kurtosis; is the expectation of the fourth power of the sample weighted center distance of the output value sample; is the sample variance of the output value sample the square of n is the number of output values ​​in the output value sample; is the sample mean of the output value samples; is the first of the output value samples i output values; is the first of the output value samples i output values The corresponding weight.

[0135] Optionally, the anomaly detection module includes: a condition judgment unit, configured to determine whether each output value sample satisfies an abnormal condition based on the adjusted kurtosis corresponding to each output value sample; The abnormal device determining unit is configured to determine the abnormal voltage detecting device when the abnormal condition is met.

[0136] Optionally, the abnormal device determination unit is configured to: When the abnormal condition is not satisfied, it is determined that no abnormal voltage detecting device exists among the respective voltage detecting devices.

[0137] Optionally, the condition judgment unit includes a first condition judgment unit, configured to: determining, based on the adjusted kurtosis of each output value sample, a standard deviation ratio of a first standard deviation of all adjusted kurtosis to a second standard deviation of all non-maximum adjusted kurtosis; When the standard deviation ratio is greater than a standard deviation ratio threshold, it is determined that each of the output value samples meets an abnormal condition.

[0138] Optionally, the condition judgment unit includes a second condition judgment unit, configured to: determining, based on the adjusted kurtosis of each output value sample, a standard deviation ratio of a first standard deviation of all adjusted kurtosis to a second standard deviation of all non-maximum adjusted kurtosis; Determine whether each output value sample meets an abnormal condition based on the standard deviation ratio and each output value sample.

[0139] Optionally, the second condition judgment unit is configured to: When the standard deviation ratio is greater than the standard deviation ratio threshold, and among the output value samples, the square of the sample variance of the output value sample corresponding to the maximum adjusted kurtosis is less than the variance square threshold, it is determined that the output value samples meet the abnormal condition.

[0140] Optionally, the second condition judgment unit is configured to: When the standard deviation ratio is greater than a standard deviation ratio threshold, and the minimum sample capacity of each output value sample is greater than a sample capacity threshold, it is determined that each output value sample meets an abnormal condition.

[0141] Optionally, the second condition judgment unit is configured to: When the standard deviation ratio is greater than the standard deviation ratio threshold, the square of the sample variance of the output value sample corresponding to the maximum adjusted kurtosis in each output value sample is less than the variance square threshold, and the minimum sample capacity in each output value sample is greater than the sample capacity threshold, it is determined that each output value sample meets the abnormal condition.

[0142] Optionally, the abnormal device determination unit is configured to: The voltage detection device corresponding to the output value sample with the largest adjusted kurtosis is used as the abnormal voltage detection device, and the number of the battery group corresponding to the abnormal voltage detection device is output.

[0143] Reference Figure 6 , shows a schematic diagram of an electronic device provided in an embodiment of the present application, comprising: at least one processor, and a memory, wherein the memory stores a computer program that can be run on the processor, wherein the processor executes the anomaly detection method described in the embodiment when executing the computer program.

[0144] Reference Figure 7 , shows a schematic diagram of a non-volatile readable storage medium provided in an embodiment of the present application, wherein the non-volatile readable storage medium stores a computer program, wherein the computer program, when executed by a processor, performs the anomaly detection method described in the embodiment.

[0145] Reference Figure 8 , shows a schematic diagram of a computer program product provided in an embodiment of the present application, including a computer program / instruction, which implements the anomaly detection method described in the embodiment when executed by a processor.

[0146] An embodiment of the present application further provides a terminal, which is used to execute the anomaly detection method described in the embodiment.

[0147] Optionally, the terminal includes a cloud terminal and a vehicle.

[0148] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0149] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, apparatuses, or computer program products. Therefore, the embodiments of the present application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware. Furthermore, the embodiments of the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0150] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0151] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0152] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0153] Although preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic inventive concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0154] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0155] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for detecting anomalies, characterized in that: The method comprises: determining, based on the output value samples corresponding to the voltage detection devices of the respective battery packs, the adjusted kurtosis corresponding to the respective output value samples, wherein the adjusted kurtosis is used to characterize the degree of abnormality of any output value sample; An abnormal voltage detection device is determined according to the adjusted kurtosis corresponding to each output value sample.

2. The method according to claim 1, characterized in that The method further comprises: Preprocess the acquired monitoring data within the target time period to determine the target data; According to the target data, output value samples corresponding to each voltage detection device are determined.

3. The method according to claim 2, characterized in that Preprocess the acquired monitoring data within the target time period to determine the target data, including: Preprocessing the monitoring data within the target time period to determine the valid data; Among the valid data, data in which the high-voltage system is in a normal state is selected as the target data.

4. The method according to claim 3, characterized in that Preprocess the monitoring data within the target time period, including: Desensitize the monitoring data within the target time period.

5. The method according to claim 4, characterized in that Preprocessing of monitoring data within the target time period also includes: Filter the fault data in the monitoring data within the target time period and sort them by time.

6. The method according to claim 2, characterized in that Determining output value samples corresponding to each voltage detection device according to the target data, including: In the target data, the output values ​​corresponding to the respective voltage detection devices that are greater than a first threshold value and less than a second threshold value are used as the output value samples corresponding to the respective voltage detection devices.

7. The method according to claim 6, characterized in that The output value is a voltage evaluation value determined after monitoring the voltage of each battery pack.

8. The method according to any one of claims 1 to 7, characterized in that Determining, based on output value samples corresponding to voltage detection devices of respective battery packs, an adjusted kurtosis corresponding to each output value sample, including: Determining the weight corresponding to each output value in each output value sample; The adjusted kurtosis corresponding to each output value sample is determined according to the weight corresponding to each output value in each output value sample.

9. The method according to claim 8, characterized in that Determining the weight corresponding to each output value in each output value sample respectively includes: For any output value sample, the weight corresponding to each output value in the output value sample is determined according to the sample standard deviation, sample mean and sample variance of the output value sample.

10. The method according to claim 9, characterized in that The calculation formula for the weight corresponding to each output value in any output value sample is: in, is the first of the output value samples i output values The corresponding weight; is the sample standard deviation of the output value sample; is the sample mean of the output value samples; is the sample variance of the output value sample; A natural constant An exponential function with base .

11. The method according to claim 8, characterized in that Determining, according to the weights corresponding to the respective output values ​​in the respective output value samples, the adjusted kurtosis corresponding to the respective output value samples, comprising: For any output value sample, determine the expectation of the fourth power of the sample weighted center distance of the output value sample according to the weights corresponding to each output value in the output value sample; The adjusted kurtosis of the output value sample is determined according to an expectation of the fourth power of the sample weighted center distance of the output value sample and the square of the sample variance of the output value sample.

12. The method according to claim 11, characterized in that The formula for calculating the adjusted kurtosis of any output value sample is: in, is the adjusted kurtosis; is the expectation of the fourth power of the sample weighted center distance of the output value sample; is the sample variance of the output value sample the square of n is the number of output values ​​in the output value sample; is the sample mean of the output value samples; is the first of the output value samples i output values; is the first of the output value samples i output values The corresponding weight.

13. The method according to claim 1, wherein Determining an abnormal voltage detection device according to the adjusted kurtosis corresponding to each output value sample, comprising: determining whether each output value sample satisfies an abnormality condition according to the adjusted kurtosis corresponding to each output value sample; When the abnormal condition is satisfied, the abnormal voltage detection device is determined.

14. The method according to claim 13, characterized in that The method further comprises: When the abnormal condition is not satisfied, it is determined that no abnormal voltage detecting device exists among the respective voltage detecting devices.

15. The method according to claim 13, characterized in that Determining whether each output value sample satisfies an abnormality condition according to the adjusted kurtosis corresponding to each output value sample includes: determining, based on the adjusted kurtosis of each output value sample, a standard deviation ratio of a first standard deviation of all adjusted kurtosis to a second standard deviation of all non-maximum adjusted kurtosis; When the standard deviation ratio is greater than a standard deviation ratio threshold, it is determined that each of the output value samples meets an abnormal condition.

16. The method according to claim 13, characterized in that Determining whether each output value sample satisfies an abnormality condition according to the adjusted kurtosis corresponding to each output value sample includes: determining, based on the adjusted kurtosis of each output value sample, a standard deviation ratio of a first standard deviation of all adjusted kurtosis to a second standard deviation of all non-maximum adjusted kurtosis; Determine whether each output value sample meets an abnormal condition based on the standard deviation ratio and each output value sample.

17. The method according to claim 16, characterized in that Determining whether each output value sample meets an abnormal condition according to the standard deviation ratio and each output value sample includes: When the standard deviation ratio is greater than the standard deviation ratio threshold, and among the output value samples, the square of the sample variance of the output value sample corresponding to the maximum adjusted kurtosis is less than the variance square threshold, it is determined that the output value samples meet the abnormal condition.

18. The method according to claim 16, characterized in that Determining whether each output value sample meets an abnormal condition according to the standard deviation ratio and each output value sample includes: When the standard deviation ratio is greater than a standard deviation ratio threshold, and the minimum sample capacity of each output value sample is greater than a sample capacity threshold, it is determined that each output value sample meets an abnormal condition.

19. The method according to claim 16, wherein Determining whether each output value sample meets an abnormal condition according to the standard deviation ratio and each output value sample includes: When the standard deviation ratio is greater than the standard deviation ratio threshold, the square of the sample variance of the output value sample corresponding to the maximum adjusted kurtosis in each output value sample is less than the variance square threshold, and the minimum sample capacity in each output value sample is greater than the sample capacity threshold, it is determined that each output value sample meets the abnormal condition.

20. The method according to any one of claims 13 to 19, characterized in that: When the abnormal condition is met, determining the abnormal voltage detection device includes: The voltage detection device corresponding to the output value sample with the largest adjusted kurtosis is used as the abnormal voltage detection device, and the number of the battery group corresponding to the abnormal voltage detection device is output.

21. An electronic device, characterized in that: include: At least one processor, and a memory, wherein the memory stores a computer program that can be run on the processor, wherein the processor executes the anomaly detection method according to any one of claims 1 to 20 when executing the computer program.

22. A non-volatile readable storage medium, characterized in that: The non-volatile readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the abnormality detection method according to any one of claims 1 to 20 is executed.

23. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the anomaly detection method according to any one of claims 1 to 20 is implemented.

24. A terminal, characterized in that: The terminal is used to execute the anomaly detection method described in any one of claims 1-20.

25. The terminal according to claim 24, characterized in that The terminal includes a cloud terminal and a vehicle.

Images