Battery insulation abnormity detection method, detection device and electronic equipment

Abstract

The invention relates to a battery insulation abnormity detection method and device and electronic equipment, and the method comprises the steps: carrying out the detection of the insulation state of a battery according to the data distribution state of insulation data, i.e., comparing the first insulation data of the battery in a first time period with the data distribution state of the historical insulation data of the battery in a historical time period; the abnormal state of the battery in the first time period is determined, and then the insulation abnormal condition of the battery is determined according to the multiple continuous battery insulation abnormal states in the first time period. According to the method provided by the invention, judgment is performed in combination with the data distribution state of the historical insulation data, the accuracy is high, the robustness is strong, misjudgment caused by a small amount of data fluctuation is avoided, and abnormity early warning can be performed in advance under the condition that the insulation abnormity of the battery exists but the safety is not influenced. According to the method provided by the invention, the insulation abnormality of the battery is determined under the condition that the abnormality exists in the plurality of continuous first time periods, the misjudgment is low, and the reliability is high.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery insulation anomaly detection, and in particular to a battery insulation anomaly detection method, a detection device and an electronic device. BACKGROUND

[0002] As a core component of new energy vehicles, the safety and stability of the battery are the key to reliable driving of the new energy vehicles. In the case of wire harness water ingress, cell liquid leakage, seal failure and other conditions in the battery, the battery may produce insulation anomalies, which may further cause vehicle driving failure. The traditional method can determine whether the insulation state of the battery is abnormal by judging against a preset threshold. However, the traditional detection method mainly determines the detection threshold according to empirical data, and the detection accuracy is not high and the reliability is low. SUMMARY

[0003] The embodiments of the present application provide a battery insulation anomaly detection method, a detection device and an electronic device, which can improve the accuracy of battery insulation anomaly detection and have high reliability, so as to at least partially solve the above technical problems.

[0004] TECHNICAL SOLUTION: In a first aspect, the battery insulation anomaly detection method provided by the embodiments of the present application comprises: determining an abnormal state of a battery in a first time period according to a data distribution state of first insulation data of the battery in the first time period and historical insulation data of the battery in a historical time period; and determining that the battery has an insulation anomaly in the case that the battery has an abnormality in connection with a plurality of the first time periods.

[0005] Optionally, the determination of the abnormal state of the battery in the first time period according to the data distribution state of the first insulation data of the battery in the first time period and the historical insulation data of the battery in the historical time period comprises: determining a first distribution state of the first insulation data and a historical distribution state of the historical insulation data; and determining that the battery is abnormal in the first time period in the case that the first distribution state deviates from the historical distribution state in a preset direction.

[0006] Optionally, the determination of the first distribution state of the first insulation data and the historical distribution state of the historical insulation data comprises: merging the first insulation data and the historical insulation data, and sorting the merged insulation data in a preset order to obtain an insulation data set; and assigning ranks to the insulation data in the insulation data set, and determining the first distribution state of the first insulation data and the historical distribution state of the historical insulation data according to the ranks of the insulation data.

[0007] Optionally, the determining that the battery is abnormal in the first time period in the case that the first distribution state deviates from the historical distribution state in the preset direction comprises: determining a first statistical value corresponding to the first distribution state and a second statistical value corresponding to the historical distribution state; determining a hypothesis probability corresponding to the preset direction according to the first statistical value and the second statistical value, and determining that the battery is abnormal in the first time period in the case that the hypothesis probability meets a preset condition.

[0008] Optionally, the method further comprises: determining a median offset value of a first median of the first insulation data and a historical median of the historical insulation data; and the determining that the battery is abnormal in the first time period in the case that the first distribution state deviates from the historical distribution state in the preset direction comprises: determining that the battery is abnormal in the first time period in the case that the first distribution state deviates from the historical distribution state in the preset direction and the median offset value is less than a preset offset threshold.

[0009] Optionally, the preset offset threshold corresponds to an abnormal level of the battery.

[0010] Optionally, the method further comprises: obtaining original working condition data of the battery; and determining the first insulation data and the historical insulation data in the original working condition data according to a preset time window and a preset working condition.

[0011] Optionally, the original working condition data comprises vehicle running data, battery current data and battery insulation data; and the method further comprises: removing battery insulation data in which the vehicle running data and the battery current data are abnormal from the original working condition data to obtain updated working condition data; and the determining the first insulation data and the historical insulation data in the original working condition data according to the preset time window and the preset working condition comprises: determining the first insulation data and the historical insulation data in the updated working condition data according to the preset time window and the preset working condition.

[0012] In a second aspect, a battery insulation abnormality detection device is provided, which comprises: a verification module configured to determine an abnormal state of a battery in a first time period according to data distribution states of first insulation data of the battery in the first time period and historical insulation data of the battery in a historical time period; and a determination module configured to determine that the battery has an insulation abnormality in the case that the battery has an abnormality in connection with a plurality of the first time periods.

[0013] In a third aspect, an electronic device is provided and includes a memory having computer programs or instructions stored thereon, and a processor configured to execute the computer programs or instructions in the memory to implement the battery insulation abnormality detection method of the first aspect.

[0014] Beneficial effects: Compared with the prior art, the technical solution provided in the embodiments of the present application detects the battery insulation state according to the data distribution state of the insulation data, that is, compares the data distribution state of the first insulation data of the battery in the first time period with the historical insulation data of the battery in the historical time period to determine the abnormal state of the battery in the first time period, and then determines the insulation abnormality of the battery according to the battery insulation abnormality state in a plurality of consecutive first time periods. The embodiments of the present application use a non-parametric method for insulation abnormality detection, which can detect the insulation abnormality of the battery without setting a specific threshold, and can be applied to the detection of batteries with various insulation levels, and has good adaptability. The embodiments of the present application combine the data distribution state of the historical insulation data for judgment, have high accuracy and strong robustness, avoid misjudgment caused by a small amount of data fluctuation, and can perform abnormal early warning in advance when the battery has insulation abnormality but does not affect safety. The embodiments of the present application determine that the battery has insulation abnormality when the battery has abnormality in a plurality of consecutive first time periods, have low misjudgment and high reliability.

[0015] Other features and advantages of the present application will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0017] In order to more completely understand the present application and its beneficial effects, the following will be described in conjunction with the drawings, wherein the same reference numerals in the following description represent the same parts.

[0018] Figure 1 is a flowchart of a battery insulation abnormality detection method provided in the embodiments of the present application;

[0019] Figure 2 is a flowchart of another battery insulation abnormality detection method provided in the embodiments of the present application;

[0020] Figure 3 is a battery insulation early warning comparison chart provided in the embodiments of the present application;

[0021] Figure 4 is a structural schematic diagram of a battery insulation abnormality detection method provided by an embodiment of the present application;

[0022] Figure 5 is a structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the protection scope of the present application.

[0024] With the rapid growth of practicability and diversity of new energy vehicles, electric vehicle related technologies are increasingly developed. As the core component of new energy vehicles, the safety and stability of the battery are the key to the reliable driving of new energy vehicles. In the case of water ingress of wiring harness, cell leakage, sealing failure and other situations inside the battery, the battery may produce insulation abnormalities, which may further cause vehicle driving failure. Insulation detection of the battery can ensure the safety and reliability of the battery system by detecting the insulation condition in the battery, and can timely warn when detecting that the battery has an abnormality, so as to avoid safety accidents due to battery abnormalities.

[0025] The factors affecting the insulation state of the battery are complex, for example, temperature, humidity, current, voltage and other factors may all cause abnormal insulation state of the battery. The traditional method can determine whether the insulation state of the battery is abnormal by judging with a preset threshold, while the battery insulation data may change dynamically in some cases, for example, in the case of water ingress due to sealing problem or failure of the explosion-proof valve of the battery pack, the battery insulation value may gradually decrease within the normal value range, and even when the water ingress condition disappears, the insulation value of the battery will gradually recover. Therefore, the detection threshold determined by the traditional method relying on historical data and experience value cannot adapt to different vehicles, resulting in low detection accuracy and low reliability.

[0026] Therefore, the embodiments of the present application provide a battery insulation abnormality detection method, a detection device and an electronic device, which can improve the accuracy of battery insulation abnormality detection and have high reliability.

[0027] Please refer to Figure 1 , Figure 1 is a flowchart of a battery insulation abnormality detection method provided by an embodiment of the present application. As shown in Figure 1 , the battery insulation abnormality detection method provided by the embodiment of the present application includes the following steps S100 to S200.

[0028] At step S100, according to a data distribution state of the first insulation data of the battery in the first time period and historical insulation data of the battery in a historical time period, an abnormal state of the battery in the first time period is determined.

[0029] The data distribution state of the insulation data represents a distribution of insulation data values in a certain time period. According to the data distribution state of the insulation data, the characteristics of the data can be determined, and the distribution rule of the insulation data can be obtained. According to the data distribution state of the insulation data in a certain time period, the insulation abnormality of the battery is determined in the embodiments of the present application. Compared with the traditional method of directly comparing the insulation data with the detection threshold to determine the insulation abnormality, the non-parametric method is used for abnormality detection, which reduces the calculation amount of the battery insulation abnormality detection.

[0030] The first time period represents a time period in which battery insulation abnormality detection needs to be performed. The first time period includes a plurality of time points, and each time point corresponds to an insulation value. For example, the first time period can be 24 hours of the day, or 12 hours from 0 o'clock to 12 o'clock of the day, or 12 hours from 12 o'clock to 0 o'clock of the day. In the embodiments of the present application, the first time period can be 24 hours of the day.

[0031] The historical time period represents a continuous time period in history. In order to facilitate detection, the historical time period can be determined according to the first time period. For example, in the case of a first time period of 12 hours, the historical time period can be a historical continuous time period of 36 hours. In the case of a first time period of 24 hours of the day, the historical time period can be a historical time period of 5 days. In the embodiments of the present application, the first insulation data in the first time period in which battery insulation abnormality detection needs to be performed is compared with the historical insulation data in the historical time period. The judgment can be made according to the data distribution state of a large amount of historical insulation data, which has strong robustness.

[0032] The battery insulation data is data representing the insulation state of the battery in the battery system, which can be directly collected. The battery insulation represents that the internal and external electrical connection parts of the battery are in an electrically isolated state, which is usually realized by electrical insulation materials. Through the insulation of the battery, it can be ensured that the high-voltage part of the battery does not directly contact the vehicle, thereby ensuring the safety of the vehicle occupants. The insulation value in the battery insulation data has certain volatility, and the battery insulation data corresponding to different vehicles also has great difference.

[0033] When the battery insulation is abnormal, the battery insulation data does not directly jump from the normal insulation value to the abnormal insulation value interval, but can gradually decrease to a lower interval value for a period of time, or decrease to a too low interval after several sudden drops. The embodiment of the present application takes the first time period of 24 hours as the first time period, and compares the data distribution state of the first insulation data in the first time period with the historical insulation data in the historical time period. When it is determined that the first insulation data is developing in the direction of reducing insulation data compared with the historical insulation data, it is determined that the abnormal state of the battery in the first time period is abnormal.

[0034] Step S200, in the case that the battery has an abnormality in the connection of a plurality of first time periods, it is determined that the battery has an insulation abnormality.

[0035] In order to improve the reliability and robustness of detection and avoid misjudgment of the word battery insulation detection, the embodiment of the present application detects the battery insulation abnormality state in a plurality of consecutive first time periods. When it is detected that the battery insulation abnormality is detected in a plurality of consecutive first time periods, it is determined that the battery has an insulation abnormality. It can be understood that when the insulation abnormality is detected in a plurality of consecutive first time periods, the same historical insulation data can be used to improve the reliability of the insulation detection.

[0036] The embodiment of the present application uses a non-parametric method to detect insulation abnormalities, which can detect the insulation abnormalities of the battery without setting a specific threshold. It can be applied to the detection of batteries with different insulation levels, and has good adaptability. The embodiment of the present application combines the data distribution state of the historical insulation data for judgment, has high accuracy and strong robustness, avoids misjudgment caused by a small amount of data fluctuation, and can perform abnormal early warning when the battery has an insulation abnormality but does not affect safety. The embodiment of the present application determines that the battery has an insulation abnormality when the battery has an abnormality in a plurality of consecutive first time periods, has low misjudgment and high reliability.

[0037] In some embodiments, according to the data distribution state of the first insulation data of the battery in the first time period and the historical insulation data of the battery in the historical time period, the abnormal state of the battery in the first time period is determined, including: determining the first distribution state of the first insulation data, and the historical distribution state of the historical insulation data; in the case that the first distribution state deviates in a preset direction from the historical distribution state, it is determined that the battery is abnormal in the first time period.

[0038] The embodiment of the present application determines the change rule of insulation data according to the data distribution state of the insulation data, so that the change trend of the insulation data can be predicted according to the data distribution state. The data distribution state of the first insulation data is determined as a first distribution state, and the data distribution state of the historical insulation data is determined as a historical distribution state. The change rule of the first insulation data is determined according to the first distribution state, and the change rule of the historical insulation data is determined according to the historical distribution state.

[0039] The first distribution state and the historical distribution state are compared, so that whether the first insulation data has insulation abnormality can be qualitatively judged. For example, if the data change rule of the first distribution state and the historical distribution state is substantially the same, it can be considered that the change trend of the first insulation data and the historical insulation data has strong consistency, and the insulation abnormality probability of the first insulation data is small; if the data change rule of the first distribution state and the historical distribution state is obviously different, it can be considered that the change trend of the first insulation data and the historical insulation data is different, and the insulation abnormality probability of the first insulation data is large.

[0040] The embodiment of the present application judges whether the first insulation data has insulation abnormality by judging the offset of the first distribution state and the historical distribution state. Exemplarily, the first distribution state is taken as a prediction distribution state, and the historical distribution state is taken as a reference distribution state. The offset of the prediction distribution state relative to the reference distribution state can include right offset or left offset, wherein the right offset means that the prediction distribution state significantly increases relative to the reference distribution state, resulting in that the prediction distribution state offsets to the direction of large data value; correspondingly, the left offset means that the prediction distribution state significantly decreases relative to the reference distribution state, resulting in that the prediction distribution state offsets to the direction of small data value.

[0041] Since the embodiment of the present application is based on the insulation data of the battery for detection, the insulation value of the battery is not suddenly jumped from the normal state to the abnormal interval, but is a qualitative change caused by quantitative change in a period of time. Therefore, in the case that the insulation data of the battery gradually decreases relative to the normal use state, it may correspond to the insulation abnormality of the battery. Therefore, the preset direction is set to the left, that is, in the case that the first distribution state of the first insulation data offsets to the left direction of the historical distribution state of the historical insulation data, it is judged that the battery has insulation abnormality in the first time period.

[0042] The embodiment of the application adopts a non-parametric battery insulation abnormality detection method. By judging the data distribution state of the insulation data, in the case that the first distribution state of the first insulation data in the first time period is left-shifted relative to the historical distribution state of the historical insulation data in the historical time period, it is considered that the first insulation data is significantly reduced relative to the historical insulation data, and it is judged that the battery has an insulation abnormality risk. In the case that the insulation data corresponds to different reduction states, the abnormality can be judged according to the method of the embodiment of the application, which has high reliability and robustness.

[0043] In some embodiments, the first distribution state of the first insulation data and the historical distribution state of the historical insulation data are determined by: merging the first insulation data and the historical insulation data, and sorting the merged insulation data in a preset order to obtain an insulation data set; and assigning ranks to the insulation data in the insulation data set, and determining the first distribution state of the first insulation data and the historical distribution state of the historical insulation data according to the ranks of the insulation data.

[0044] As described in the foregoing embodiments, the time period includes a plurality of time points, the first insulation data in the first time period includes insulation values corresponding to the plurality of time points, and the historical insulation data in the historical time period also includes insulation values corresponding to the plurality of time points. In order to facilitate subsequent comparison according to the first distribution state and the historical distribution state, the first insulation data and the historical insulation data are assigned ranks in the embodiment of the application, so as to determine the data distribution state of the insulation data according to the ranks of each insulation value.

[0045] Before assigning the ranks to the insulation data, the first insulation data and the historical insulation data are merged, and each insulation value is sorted from small to large to obtain a merged and sorted insulation data set, and then the ranked insulation data set is assigned ranks. Assigning the ranks can be understood as marking the serial numbers of each data in the insulation data set. By assigning the ranks to the insulation data, the data distribution state of the insulation data can be further determined, and it is also convenient for subsequent judgment of the shift of the data distribution state according to the ranks.

[0046] Exemplarily, the insulation values included in the first insulation data are represented as [5, 8, 10, 12], the insulation values included in the historical insulation data are represented as [6, 7, 11, 13], the insulation data set obtained by merging and sorting the first insulation data and the historical insulation data is represented as [5, 6, 7, 8, 10, 11, 12, 13]. The rank of each insulation value in the insulation data set can be obtained as follows: the rank of the insulation value “5” is 1, the rank of the insulation value “6” is 2, the rank of the insulation value “7” is 3, …, and the rank of the insulation value “13” is 8. The sum of the ranks of all insulation values in the insulation data is the rank sum of the insulation data, for example, the rank sum of the first insulation data is 1+4+5+7=17, and the rank sum of the historical insulation data is 2+3+6+8=19. The distribution state of the insulation data can be obtained according to the ranks and the rank sum.

[0047] In some embodiments, in the case that the first distribution state deviates from the historical distribution state in a preset direction, it is determined that the battery is abnormal in the first time period, including: determining a first statistical value corresponding to the first distribution state, and a second statistical value corresponding to the historical distribution state; determining a hypothesis probability corresponding to the preset direction according to the first statistical value and the second statistical value, and determining that the battery is abnormal in the first time period in the case that the hypothesis probability meets a preset condition.

[0048] In the case that the insulation data and the rank sum of the insulation data are known, the statistical value of the insulation data can be further calculated. In the case that the rank and the insulation data are determined, the statistical value can be calculated by different calculation methods, and the calculation method of the statistical value is not limited in the embodiments of the present application, as long as the statistical value of the insulation data can be calculated according to the rank and the insulation data. Exemplarily, the first statistical value is obtained according to the first insulation data and the rank sum of the first insulation data, the second statistical value is obtained according to the historical insulation data and the rank sum of the historical insulation data, and the smaller statistical value of the first statistical value and the second statistical value is taken as the target statistical value for the deviation state detection.

[0049] According to the target statistical value, the first insulation data, the historical insulation data and the preset direction of the data deviation, a hypothesis probability is determined, which is represented as P. The hypothesis probability can be determined by a preset test statistical table, or calculated by statistical software based on an exact distribution or an approximate distribution of a large sample. After the hypothesis probability is determined, the hypothesis probability is compared with a significance level threshold α corresponding to the preset direction, if P≤α, the original hypothesis is rejected, and it is considered that there is statistical evidence supporting the alternative hypothesis under the selected significance level, that is, it is considered that the first insulation data deviates from the historical insulation data in the preset direction, and the battery is abnormal in the first time period. If P>α, it indicates that there is not enough evidence to reject the original hypothesis, and it cannot be concluded that there is a significant difference between the first insulation data and the historical insulation data.

[0050] The embodiment of the present application determines the offset state of the first insulation data in the first time period relative to the historical insulation data in the historical time period according to the data distribution state of the insulation data, and further determines the abnormal state of the battery in the first time period. The embodiment of the present application does not need to set the insulation detection threshold according to the historical data or the experience data to determine whether there is an abnormality through the judgment of the insulation detection threshold, and can improve the reliability and accuracy of the battery insulation abnormality judgment.

[0051] In some embodiments, the battery insulation abnormality detection method further comprises: determining a median offset value of a first median of the first insulation data and a historical median of the historical insulation data; in the case that the first distribution state offsets in the preset direction of the historical distribution state, determining that the battery is abnormal in the first time period, comprising: in the case that the first distribution state offsets in the preset direction of the historical distribution state and the median offset value is less than a preset offset threshold, determining that the battery is abnormal in the first time period.

[0052] The median represents the data at the middle position of the insulation data, and the median can represent the middle value of the data distribution. The embodiment of the present application obtains the median offset value according to the change ratio of the first median of the first insulation data and the historical median of the historical insulation data.

[0053] In order to improve the reliability of the battery insulation detection, the embodiment of the present application assists in judging the abnormal state of the battery insulation by calculating the median offset value of the insulation data. Only in the case that the first distribution state offsets in the preset direction of the historical distribution state, and the median offset value of the first median of the first insulation data and the historical median of the historical insulation data is less than the preset offset threshold, it can be determined that the battery is abnormal in the first time period, and the reliability and accuracy of the battery insulation abnormality judgment are improved through the comprehensive judgment of the two conditions.

[0054] In some embodiments, the preset offset threshold corresponds to the abnormal level of the battery. The preset offset threshold can be determined and set according to the abnormal level of the battery, and different abnormal levels correspond to different preset offset thresholds. While detecting the insulation state of the battery, the battery insulation abnormality can also be classified, and in the case of high-level insulation abnormality, a warning and high-priority abnormality processing are performed.

[0055] For example, the preset offset threshold can be -20%, -40%, and -60%, which respectively correspond to a first abnormality, a second abnormality, and a third abnormality. If the current median offset value is greater than the preset offset threshold of the first abnormality, it is judged that the insulation data is normal, and if the current median offset value is less than or equal to the preset offset threshold of the first abnormality, it is judged that the insulation data is abnormal, and it is considered that the current insulation data has a first abnormality.

[0056] In some embodiments, the battery insulation anomaly detection method further comprises: obtaining original working condition data of the battery; determining first insulation data and historical insulation data in the original working condition data according to a preset time window and a preset working condition.

[0057] The original working condition data of the battery can be obtained from a big data platform of the power battery. The original working condition data of the battery represents various running data and state data collected during use of the battery. The insulation data of the battery can also be obtained from the original working condition data. The original working condition data includes data of the battery under different working conditions, such as a charging working condition, corresponding to a working condition of the battery when the vehicle is stationary; a discharging working condition, corresponding to a working condition of the vehicle in a driving state; and a stationary working condition, corresponding to a working condition of the battery when the vehicle is stationary.

[0058] Since different charging equipment under the charging working condition can cause insulation anomalies of the vehicle battery, and the data volume under the discharging working condition is greater than that under the charging working condition, the accuracy of the battery insulation anomaly detection can be improved. Therefore, the preset working condition is set to the discharging working condition in the embodiments, that is, the insulation data under the discharging working condition is selected from the original working condition data as the data for battery insulation anomaly detection.

[0059] After the preset working condition is determined, the first insulation data corresponding to a first time and the historical insulation data corresponding to a historical time period are extracted according to the time window. It can be understood that the larger the time window for extracting the historical insulation data, the higher the calculation accuracy, but the lower the calculation efficiency. The embodiments of the present application select five days as the time window, which can balance the calculation accuracy and the calculation efficiency.

[0060] In some embodiments, the original working condition data includes vehicle running data, battery current data, and battery insulation data; the battery insulation anomaly detection method further comprises: removing the battery insulation data with abnormal vehicle running data and battery current data from the original working condition data to obtain updated working condition data; determining the first insulation data and the historical insulation data in the original working condition data according to the preset time window and the preset working condition, including: determining the first insulation data and the historical insulation data in the updated working condition data according to the preset time window and the preset working condition.

[0061] In some cases, there are abnormal insulation data in the original working condition data. For example, in the case that the insulation data cannot be detected or the detection is invalid, the insulation data may be defaulted as an extremely large or extremely small abnormal value. The above-mentioned abnormal battery insulation data needs to be removed from the original working condition data to improve the accuracy and reliability of the battery abnormality detection. In addition, under certain conditions of vehicle running data or battery current data, the battery insulation data may be distorted, affecting the accuracy of the battery insulation detection. The embodiments of the present application can also filter out the battery insulation data under the abnormal conditions of vehicle running data or battery current data through a sliding window to improve the accuracy and reliability of the battery insulation abnormality detection.

[0062] The embodiments of the present application determine whether the first insulation data deviates significantly from the historical insulation data based on the data distribution state of the insulation data, without setting a detection threshold according to historical data or expert experience. The detection method is simple and practical and has stronger robustness.

[0063] The embodiments of the present application also assist in judging the battery insulation abnormality by combining the median deviation value. Through the double conditions of non-parametric, the abnormal downward trend of insulation sudden drop and sustained slow drop can be effectively identified, and early warning can be given before the battery insulation abnormality affects use and causes safety problems. The embodiments of the present application quantify the risk of battery insulation abnormality by the deviation value of the median and the abnormality level division, avoid misjudgment caused by normal deviation of insulation, and can adapt to vehicle batteries with different insulation levels, which is more practical.

[0064] The battery insulation abnormality detection method of the embodiments of the present application will be described below through a specific example. Please refer to Figure 2 , Figure 2 is a flow diagram of another battery insulation abnormality detection method provided in the embodiments of the present application. As shown in Figure 2 , the battery insulation abnormality detection method includes the following steps S210 to S260.

[0065] Step S210, obtaining original working condition data of a battery on a big data platform of a power battery.

[0066] The original working condition data of the battery includes vehicle running data, battery insulation data, battery current data, time data, etc. The working conditions in the original working condition data of the battery include data in discharge condition, charging condition, and standing condition.

[0067] Step S220, extracting working condition data in the discharge condition from the original working condition data.

[0068] Step S230, filtering abnormal values from the working condition data in the discharge condition.

[0069] The abnormal value includes invalid value of insulation data, and insulation data when the battery current data is less than 5A continuously.

[0070] In step S240, the insulation data of the day and the historical insulation data are determined in the filtered charging working condition data.

[0071] The insulation data in the time period of the day is taken as the insulation data of the day according to the time window of the day, and is recorded as data_ir_dt, and the insulation data of the past five days is taken as the historical insulation data according to the historical time window, and is recorded as data_ir_history. The size of the historical window can be set by itself, and the larger the window is, the more accurate the data distribution is, but the larger the window is, the lower the calculation efficiency is.

[0072] In step S250, the insulation abnormal state of the battery of the day is determined according to the first distribution state of the insulation data of the day and the historical distribution state of the historical insulation data.

[0073] (a) The insulation data of the day and the historical insulation data are combined and sorted.

[0074] The insulation data of the day data_ir_dt and the historical insulation data data_ir_history are combined into an insulation data set, and the insulation values in the insulation data set are sorted from small to large according to the size.

[0075] (b) The data in the insulation data set is assigned a rank.

[0076] A unique rank is assigned to each sorted insulation value in the insulation data set, and the rank of the insulation value is the serial number of the insulation value in the insulation data set. If there are the same insulation values in the insulation data set, the same rank is assigned to the same insulation values, and the common rank is determined according to the arithmetic average of the position serial number of the insulation value.

[0077] (c) The sum of the ranks is calculated.

[0078] The sum of the ranks corresponding to the insulation data of the day data_ir_dt and the historical insulation data data_ir_history is calculated respectively.

[0079] (d) The U statistic is calculated.

[0080] Based on the insulation data of the day data_ir_dt and the historical insulation data data_ir_history and the sum of the ranks, the U statistic of the insulation data is calculated respectively, and the U statistic can be used to determine the difference between the insulation data of the day and the historical insulation data. The U statistic corresponding to the insulation data of the day is recorded as the first U statistic, and the U statistic corresponding to the historical insulation data is recorded as the second U statistic, and the smaller U statistic between the first U statistic and the second U statistic is selected as the final target statistic.

[0081] (e) determining the hypothesis probability.

[0082] According to the finally calculated target statistic, the current insulation data data_ir_dt, the historical insulation data data_ir_history and the preset hypothesis direction, i.e. the current insulation data data_ir_dt is generally offset to the left of the historical insulation data data_ir_history, the corresponding hypothesis probability, denoted as P value, is calculated or obtained. The hypothesis probability can be obtained by consulting a special U test statistical table or automatically calculated by statistical software based on the exact distribution or large sample approximate distribution.

[0083] (f) making a decision based on the P value.

[0084] The calculated P value is compared with the pre-selected significance level a. For example, in the case of a = 0.05, if P < a, the null hypothesis is rejected, and it is considered that there is statistical evidence to support the alternative hypothesis that the data distribution of the current insulation data is statistically different from the data distribution of the historical insulation data, i.e. the current insulation data data_ir_dt is generally distributed to the left of the historical insulation data data_ir_history. If P > a, there is not enough evidence to reject the null hypothesis, and it cannot be concluded that there is a significant difference.

[0085] Step S260, determining the insulation abnormal state of the battery on the current day according to the median offset value between the first median of the current insulation data and the historical median of the historical insulation data.

[0086] The first median ir_median_dt of the current insulation data data_ir_dt and the historical median ir_median_history of the historical insulation data data_ir_history are calculated respectively. The median offset value ir_ratio of the first median and the historical median is calculated, and the calculation expression can be:

[0087] ir_ratio = (ir_median_dt - ir_median_history) / ir_median_history * 100%

[0088] If the median offset value ir_ratio is less than the preset offset value -20%, it is judged that the current insulation data has insulation abnormality. The preset offset value can be set to multiple different values, such as -20%, -40%, -60%, which correspond to 1st, 2nd and 3rd insulation abnormality levels respectively.

[0089] Step S270, determining the insulation abnormal state of the battery on the current day.

[0090] In the case that the insulation abnormality of the battery is detected in step S250 and step S260 simultaneously, it is determined that the insulation abnormality of the battery exists on the current day.

[0091] In step S280, it is determined that the abnormal state of the battery exists for consecutive days.

[0092] In the case that the insulation abnormality of the battery exists for consecutive days, it is determined that the insulation abnormality of the battery exists, and the abnormality early warning is performed. In the abnormality early warning, the insulation abnormality level of the battery can be determined according to the preset offset threshold value in the determination of the median offset value.

[0093] Please refer to Figure 3 , Figure 3 is a battery insulation early warning comparison chart provided in the embodiments of the present application. As shown in Figure 3 , the red points are the dates when the insulation abnormality of the battery is determined, the pink area is the normal descending level area, and the blue is the date when the insulation is determined to be normal. Figure 3 (a) shows the insulation data sudden drop schematic diagram, and in Figure 3 (a), the insulation drops suddenly but does not reach the low standard, and the battery can still be used normally; Figure 3 (b) shows the insulation data slow decline schematic diagram, and the battery is in the normal interval in the slow decline process. Both of the two vehicles are in the last day in the vehicle end to produce the early warning due to the low insulation data, and according to the battery insulation prediction method of the embodiments of the present application, the insulation abnormality of the battery is detected before the early warning in the vehicle end.

[0094] The embodiments of the present application evaluate whether the distribution position of the current day insulation data and the data in the historical window has deviated by using the data distribution state detection method of one-sided U test, and then determine whether the current day insulation data has appeared the decline phenomenon. This non-parametric method does not need the historical data or the experience threshold value set by the expert, and can adapt to the power battery of different insulation levels; the embodiments of the present application refer to the distribution of a large amount of insulation data, which has higher robustness for the strongly fluctuating insulation data, and avoids the misjudgment caused by a small number of fluctuation points. At the same time, the embodiments of the present application also assist in judging the battery insulation abnormality by calculating the offset value of the median, provide a quantifiable reference for the abnormality level of the battery insulation abnormality, and also avoid the one-sided U test misjudgment caused by the normal deviation of the insulation data by setting the lowest abnormality threshold value.

[0095] The embodiments of the present application can judge the insulation abnormality of the battery before the insulation abnormality of the battery reaches the extremely low level which does not affect the normal use or has not yet caused the safety problem, and perform the early warning before the substantial safety problem is caused. In order to further increase the reliability of the battery abnormality detection, the embodiments of the present application also determine whether the insulation abnormality is detected for consecutive N days by using the historical detection results, prevent the false early warning caused by the single day misjudgment, and have high reliability.

[0096] Based on the same inventive concept, the embodiment of the present application also provides a battery insulation abnormality detection device for implementing the battery insulation abnormality detection method. The device provides a solution to the implementation scheme as described in the above method, and therefore the specific limitations in one or more battery insulation abnormality detection device embodiments provided below can refer to the limitations of the battery insulation abnormality detection method described above, and will not be repeated here.

[0097] In some embodiments, referring to Figure 4 , Figure 4 is a structural schematic diagram of a battery insulation abnormality detection device provided in the embodiment of the present application. As Figure 4 indicated, the battery insulation abnormality detection device provided in the embodiment of the present application includes a verification module 410 and a determination module 420; wherein:

[0098] The verification module is configured to determine the abnormal state of the battery in the first time period according to the data distribution state of the first insulation data of the battery in the first time period and the historical insulation data of the battery in the historical time period.

[0099] The determination module is configured to determine that the battery has insulation abnormality in the case that the battery has abnormality in the connection of the plurality of first time periods.

[0100] Each module in the above battery insulation abnormality detection device can be realized by software, hardware and their combination in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to call and execute the operations corresponding to the above modules by the processor.

[0101] Figure 5 is a structural schematic diagram of an electronic device provided in the embodiment of the present application. Correspondingly, the embodiment of the present application also provides an electronic device, referring to Figure 5 , the electronic device includes a memory, a processor and a computer program or instructions stored in the memory and executable on the processor, and the processor executes the computer program or instructions to implement the steps of the above battery insulation abnormality detection method. Since the battery insulation abnormality detection method is described in detail above, it will not be repeated here.

[0102] Correspondingly, the embodiment of the present application also provides a computer readable storage medium having a computer program or instructions stored thereon, and the computer program or instructions are executed by the processor to implement the steps of the above battery insulation abnormality detection method. Since the battery insulation abnormality detection method is described in detail above, it will not be repeated here.

[0103] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0104] Those skilled in the art will appreciate that the drawings are merely schematic diagrams of exemplary embodiments and may not be to scale. The modules or processes in the drawings are not necessarily required to implement the present application and therefore cannot be used to limit the scope of protection of the present application.

[0105] The above is a detailed introduction to the battery insulation abnormality detection method, detection device and electronic device provided in the embodiments of the present application, and specific examples are used to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the technical solution and core idea of ​​the present application; ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents; and these modifications or replacements do not cause the essence of the corresponding technical solution to deviate from the scope of the technical solution of the embodiments of the present application.

Claims

1. A battery insulation abnormality detection method, characterized in that: include: determining an abnormal state of the battery in the first time period according to first insulation data of the battery in the first time period and data distribution states of historical insulation data of the battery in historical time periods; If the battery has an abnormality during the connection of a plurality of the first time periods, it is determined that the battery has an insulation abnormality.

2. The method according to claim 1, characterized in that Determining an abnormal state of the battery in the first time period according to first insulation data of the battery in the first time period and data distribution states of historical insulation data of the battery in historical time periods includes: determining a first distribution state of the first insulation data and a historical distribution state of the historical insulation data; When the first distribution state deviates toward a preset direction of the historical distribution state, it is determined that the battery is abnormal during the first time period.

3. The method according to claim 2, characterized in that The determining of the first distribution state of the first insulation data and the historical distribution state of the historical insulation data includes: Merging the first insulation data and the historical insulation data, and sorting the merged insulation data according to a preset order to obtain an insulation data set; The insulation data in the insulation data set are ranked, and a first distribution state of the first insulation data and a historical distribution state of the historical insulation data are determined according to the ranking of the insulation data.

4. The method according to claim 3, characterized in that When the first distribution state deviates in a preset direction toward the historical distribution state, determining that the battery is abnormal during the first time period includes: Determine a first statistical value corresponding to the first distribution state and a second statistical value corresponding to the historical distribution state; A hypothetical probability corresponding to the preset direction is determined according to the first statistical value and the second statistical value, and when the hypothetical probability meets a preset condition, it is determined that the battery is abnormal during the first time period.

5. The method according to claim 2, characterized in that The method further comprises: determining a median offset value between a first median of the first insulation data and a historical median of the historical insulation data; When the first distribution state deviates in a preset direction toward the historical distribution state, determining that the battery is abnormal during the first time period includes: When the first distribution state shifts toward a preset direction of the historical distribution state and the median shift value is less than a preset shift threshold, it is determined that the battery is abnormal during the first time period.

6. The method according to claim 5, characterized in that The preset offset threshold corresponds to an abnormality level of the battery.

7. The method according to claim 1, characterized in that The method further comprises: Obtaining original operating condition data of the battery; The first insulation data and the historical insulation data are determined in the original operating condition data according to a preset time window and a preset operating condition.

8. The method according to claim 7, characterized in that The original operating condition data includes vehicle operation data, battery current data, and battery insulation data; the method further includes: Eliminating the vehicle operation data and battery insulation data with abnormal battery current data from the original operating condition data to obtain updated operating condition data; The determining the first insulation data and the historical insulation data in the original operating condition data according to a preset time window and a preset operating condition includes: The first insulation data and the historical insulation data are determined in the updated operating condition data according to a preset time window and a preset operating condition.

9. A battery insulation abnormality detection device, characterized in that: include: a testing module configured to: determine an abnormal state of the battery in the first time period according to first insulation data of the battery in the first time period and data distribution states of historical insulation data of the battery in historical time periods; The determination module is configured to: determine that the battery has an insulation abnormality when the battery has an abnormality during a plurality of the first time periods.

10. An electronic device, characterized in that: include: Memory on which computer programs or instructions are stored; A processor, configured to execute the computer program or instructions in the memory to implement the battery insulation abnormality detection method according to any one of claims 1 to 8.

Images