Method and System for Monitoring the Operating State of a Battery Automatic Short-Circuit and Reverse-Polarity Detection Device

Abstract

The present invention relates to the field of data processing. More specifically, the present invention relates to a method and system for monitoring the operating state of a battery automatic short-circuit reverse-polarity detection device. The method includes: obtaining the historical data of current, voltage, and temperature of the battery detection device, calculating the degree of data fluctuation using the HWKS algorithm, setting an initial k' value, clustering the historical operating data at different levels, analyzing the degree of aggregation and local density of the clustering clusters, correcting the initial k' value to obtain a corrected significant k value, using the significant k value as the parameter value in the HWKS algorithm, obtaining the degree of abnormality of each level of data, performing a weighted sum of the degrees of abnormality of each level of data to obtain the comprehensive degree of abnormality of each level of data, determining whether a data point is an abnormal data based on the comprehensive degree of abnormality, and monitoring the detection device. The present invention accurately identifies abnormal points in the data by adaptively adjusting the k value, thereby improving the accuracy of operation monitoring.

Description

Technical Field

[0001] The present invention relates to the field of data processing. More specifically, the present invention relates to a method and system for monitoring the operating state of a battery automatic short - circuit reverse - polarity detection device. Background Art

[0002] Batteries play an important role in modern technology and are widely used in various devices, such as electric vehicles, portable electronic devices, and energy storage systems. The performance and safety of batteries directly affect the reliability of devices and the safety of users. Currently, when monitoring detection devices, it involves the detection of multi - level data. The detection result of a single piece of data is relatively single and cannot accurately reflect the operating state of the detection device.

[0003] The existing Chinese patent application document with the publication number CN118518948A discloses an electromagnetic compatibility detection method, device, equipment, and storage medium. The electromagnetic compatibility detection method includes: obtaining multi - dimensional information of a detection device, and based on the multi - dimensional information of the detection device, setting a multi - level test range of electromagnetic environment intensity; according to the set multi - level test range of electromagnetic environment intensity, collecting operating state data of the detection device to obtain the operating state data of the detection device under each level of electromagnetic environment intensity; performing stability calibration on the operating state data to obtain calibrated operating state data, and performing dynamic serialization processing on the calibrated operating state data to obtain dynamic sequence operating state data. This invention can predict potential faults through the monitoring and classification of device states, and take maintenance or repair measures in a timely manner, extending the service life of the device and improving the reliability of the system.

[0004] In the above - mentioned application document, multi - dimensional information of the detection device is obtained, stability calibration and dynamic serialization processing are performed on the operating state data, and classification processing of a machine learning algorithm model is carried out to obtain normal or abnormal operating data. Currently, in combination with the HWKS algorithm, since the value is set manually or is an empirical value, for areas with a relatively dense data point, a smaller value can provide more refined detection; in areas with a relatively sparse data point, a larger value can provide more context information. Since there are also certain differences in the fluctuation degrees of the detection device at different levels during operation, at this time, a fixed value cannot adapt to the differences of data at each level, and thus there is a certain deviation in the final abnormal detection result of the data point, affecting the accuracy of the result. Summary of the Invention

[0005] To solve the problem that a fixed value cannot adapt to the differences of data at each level, resulting in a certain deviation in the final abnormal detection result of the data point, the present invention provides solutions in the following aspects.

[0006] In the first aspect, a method for monitoring the operating state of a battery automatic short-circuit and reverse-polarity detection device includes: obtaining historical operating data at each level of the battery automatic short-circuit and reverse-polarity detection device, where the historical operating data at each level includes: current, voltage, and temperature; based on the HWKS algorithm, calculating the historical change rate and change frequency of the historical operating data at the corresponding level to calculate the fluctuation degree of the historical operating data at the corresponding level, and using the fluctuation degree as the initial value; clustering the historical operating data at each level to obtain multiple clustering clusters, obtaining the aggregation degree between each clustering cluster, and calculating the local density of the clustering cluster according to the distance difference and aggregation degree difference between the clustering centers of each clustering cluster; modifying the initial value according to the local density to obtain a modified significant value, using the significant value as the parameter value in the HWKS algorithm, obtaining the abnormality degree of each level of data, performing weighted summation on the abnormality degrees of each level of data to obtain the comprehensive abnormality degree of each level of data, and judging whether the data point is abnormal data based on the comprehensive abnormality degree, and monitoring the detection device.

[0007] The effect is that: through effective monitoring and timely fault handling, the damage to the battery can be reduced, thereby prolonging the service life of the battery, being able to adapt to different operating conditions and environmental changes, dynamically calculating the initial value and the significant value using the HWKS algorithm, enabling the detection system to adapt to data changes, and improving the flexibility and robustness of the detection system.

[0008] Preferably, the initial value includes the steps of:

[0009] Selecting historical data within a preset time period at the data point acquisition moment in the historical data, plotting a line graph for the selected historical data, obtaining the extreme points in the line graph, calculating the slope between adjacent extreme points, and forming a slope set;

[0010] Taking the square of the difference between the sum of the absolute values of the positive slopes and the sum of the absolute values of the negative slopes as the slope difference, taking the square of the difference in the probabilities of the positive slopes and the negative slopes as the probability difference, and taking the square root of the sum of the slope difference and the probability difference to evaluate the fluctuation degree of each level of data, and using the fluctuation degree as the initial value.

[0011] The effect is that: by calculating the slope difference and the probability difference, the volatility of the data within the preset time period can be quantified, providing a more suitable Value; By distinguishing between positive and negative slopes, the directionality of data changes is considered, which helps to identify anomalies in the data. Furthermore, when the data fluctuates greatly, by adjusting the value, the sensitivity of the detection system can be improved, and potential anomalies can be captured better.

[0012] Preferably, obtaining the degree of aggregation between each clustering cluster includes:

[0013] Select historical data in a preset time period before the data point collection time in the historical data, cluster the selected local data to obtain several clustering clusters, take any clustering cluster as the target clustering cluster, and obtain the average distance between each data point in the target clustering cluster and the clustering center of the target clustering cluster as the degree of aggregation.

[0014] The effect is that by calculating the average distance between the data points in the clustering cluster and the clustering center, the smaller the average value indicates that the data points are more closely clustered around the clustering center, and the degree of aggregation of the data in the local area can be captured.

[0015] Preferably, the local density of the clustering cluster includes:

[0016] Take any clustering cluster as the target clustering cluster, obtain the reciprocal of the distance between the clustering center of the target clustering cluster and the clustering center of the neighboring clustering cluster, and the reciprocal of the difference between the degree of aggregation of the target clustering cluster and the degree of aggregation of the neighboring clustering cluster;

[0017] Take the reciprocal of the distance between the clustering centers as the weight of the reciprocal of the difference between the degrees of aggregation, and perform weighted summation as the local density of the clustering cluster.

[0018] The effect is that by taking the reciprocal of the distance between the clustering centers as the weight, the influence of neighboring clustering clusters on the local density is emphasized, making the calculation of the local density pay more attention to the data points in the neighboring area. By considering the local density, the anomaly detection ability can be enhanced to improve the accuracy of anomaly detection.

[0019] Preferably, the significant value satisfies the following relational expression:

[0020] ;

[0021] In the formula, represents the significant value of each level of data, represents a preset hyperparameter, represents the local density of the target clustering cluster, represents the initial value of each level of data, represents as the base exponential function.

[0022] Its effects are as follows: By significantly The dynamic adjustment of the value helps to improve the accuracy of anomaly detection, can be optimized according to the local characteristics of the data, and at the same time significantly The dynamic adjustment of the value can reduce the influence of outliers and noise, thereby enhancing the robustness of the algorithm on different data sets.

[0023] Preferably, the comprehensive anomaly degree satisfies the following relational expression:

[0024] ;

[0025] In the formula, represents the comprehensive anomaly degree, represents the anomaly degree of the voltage level, represents the anomaly degree of the current level, represents the anomaly degree of the temperature level, respectively represent the voltage level weight, the current level weight, and the temperature level weight.

[0026] Preferably, based on the comprehensive anomaly degree, it is judged whether the data point is abnormal data, and the detection device is monitored, including:

[0027] In response to the comprehensive anomaly degree being greater than the anomaly threshold, it is abnormal data; otherwise, it is normal data. After obtaining the abnormal data, the warning device is triggered to monitor the detection device in real time.

[0028] In the second aspect, a monitoring system for the operating state of a battery automatic short-circuit and reverse-polarity detection device includes: a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned method for monitoring the operating state of the battery automatic short-circuit and reverse-polarity detection device is implemented.

[0029] The present invention has the following effects:

[0030] 1. By adaptively adjusting the value, the algorithm can dynamically adjust according to the local density and hierarchical fluctuations of the data points, thereby improving the accuracy of anomaly detection, and can more accurately identify the abnormal points in the data, reducing the situations of false detection and missed detection.

[0031] 2. Based on the fluctuation degree and local density of the data points at different levels, the present invention can significantly improve the monitoring effect of the HWKS algorithm on the operating state of the battery automatic short-circuit and reverse-polarity detection device. It not only improves the accuracy and robustness of anomaly detection, but also enhances the sensitivity and adaptability to the changes in the device state. Description of the Drawings

[0032] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understandable. In the drawings, several embodiments of the present invention are shown by way of example and not limitation, and like or corresponding reference numerals represent like or corresponding parts, wherein:

[0033] Figure 1 is a flowchart of the method for steps S1 - S4 in the method for monitoring the operating state of the battery automatic short - circuit and reverse - polarity detection device according to an embodiment of the present invention.

[0034] Figure 2 is a block diagram of the structure of the system for monitoring the operating state of the battery automatic short - circuit and reverse - polarity detection device according to an embodiment of the present invention. Detailed Embodiments

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present invention.

[0036] The following will describe in detail the specific embodiments of the present invention with reference to the accompanying drawings.

[0037] Referring to Figure 1 , the method for monitoring the operating state of the battery automatic short - circuit and reverse - polarity detection device includes steps S1 - S4, specifically as follows:

[0038] S1: Obtain the historical operation data of each level of the battery automatic short - circuit and reverse - polarity detection device, where the historical operation data of each level includes: current, voltage, and temperature.

[0039] Further explanation, in this embodiment, by installing sensors in the battery automatic short - circuit and reverse - polarity detection device (hereinafter referred to as: detection device), the current, voltage, temperature, etc. of the device are collected. The historical operation data includes, but is not limited to: current , voltage , temperature , etc. Among them, due to the different dimensions of the historical operation parameters, the historical operation data is stratified. The data in the current dimension is the current layer, the data in the voltage dimension is the voltage layer, and the data in the temperature dimension is the temperature layer. Anomaly detection is performed on the data of each level respectively.

[0040] It should be noted that during the process of monitoring the operating state of the detection device, the operating parameters of the detection device belong to multi - dimensional data. Furthermore, using the HWKS algorithm, anomaly detection can be performed at different levels. The HWKS algorithm refers to an algorithm based on a hierarchical structure and weighted The K-Nearest Neighbors (K-NN) algorithm is used for anomaly detection or classification tasks. This algorithm mainly obtains the anomaly performance of a data point at a certain level by calculating the similarity between the data point and the data points within the neighborhood range at different levels.

[0041] S2: Based on the HWKS algorithm, the historical change rate and change frequency of historical operation data at its corresponding level are used to calculate the fluctuation degree of historical operation data at its corresponding level, and the fluctuation degree is used as the initial value of the HWKS algorithm.

[0042] Select the historical data within a preset time period at the data point collection moment from the historical data, draw a line chart for the selected historical data, obtain the extreme points in the line chart, and calculate the slopes between adjacent extreme points to form a slope set;

[0043] Specifically, the slope set satisfies the following relational expression:

[0044] ;

[0045] In the formula, represents the slope set of the current layer, represents the first extreme point of the line chart of the current layer in the historical data, represents the second extreme point of the line chart of the current layer in the historical data, represents the third extreme point of the line chart of the current layer in the historical data, represents the th extreme point of the line chart of the current layer in the historical data, represents the th extreme point of the line chart of the current layer in the historical data, represents the slope of the line passing through the second extreme point and the first extreme point, represents the slope of the line passing through the third extreme point and the second extreme point, represents the th extreme point and the th extreme point of the line chart of the current layer in the historical data.

[0046] Further explanation, in this embodiment, the preset time period is one hour before the data point collection moment as a time period, and the extreme points in the line chart are local maximum and minimum values. Specifically, the slope between adjacent extreme points is the slope of the line formed by the two extreme points.

[0047] The square of the difference between the sum of the absolute values of the positive slopes and the sum of the absolute values of the negative slopes is taken as the slope difference, and the square of the difference in the probabilities of the occurrence of the positive slopes and the negative slopes is taken as the probability difference. The square root of the sum of the slope difference and the probability difference is used to evaluate the degree of fluctuation of each level of data, and the degree of fluctuation is used as the initial value.

[0048] Specifically, the initial value satisfies the following relational expression:

[0049] ;

[0050] In the formula, represents the degree of fluctuation of each level of data, represents the cumulative sum of the positive slopes in the slope concentration of the current layer, represents the cumulative sum of the negative slopes in the slope concentration of the current layer, represents the probability of the positive slopes in the slope concentration of the current layer, represents the probability of the negative slopes in the slope concentration of the current layer.

[0051] Further explanation, taking the current layer data as an example, obtaining the slope set of the current layer data. According to the fact that the larger the sum of the squares of the two differences in the slope concentration, the greater the degree of fluctuation of the slope set of the current layer, and the greater the degree of fluctuation of the current layer of the historical data corresponding to the data points at the acquisition moment. At this time, the initial value of this data point is smaller, so as to reduce the influence of this kind of fluctuation on the result accuracy of the abnormal performance of the data point in the current layer. That is to say, the difference in the sum of the absolute values of the positive slopes and the negative slopes is large, and the probability difference is also large, indicating that the volatility of the slopes of the current layer is very high, and the change amplitude and frequency between rising and falling of the data are unbalanced, resulting in a greater degree of fluctuation.

[0052] S3: Cluster the historical operation data of each level to obtain multiple clustering clusters, and obtain the degree of aggregation between each clustering cluster. According to the distance difference and the degree of aggregation difference between the clustering centers of each clustering cluster, it is used to calculate the local density of the clustering cluster.

[0053] Select the historical data in the preset period before the data point acquisition moment in the historical data, cluster the selected local data to obtain several clustering clusters, take any clustering cluster as the target clustering cluster, and obtain the average value of the distances between each data point in the target clustering cluster and the clustering center of the target clustering cluster as the degree of aggregation;

[0054] Further explanation: The preset time period is the historical data in the previous 1 minute before the data point collection moment. Exemplarily, in the clustering result, the aggregation degree of the clustering clusters and the distance between the clustering clusters reflect the local density of the data point at this level. The higher the aggregation degree of the clustering clusters and the closer the distance between the clustering clusters, the higher the density of this part of the data. Taking the process of obtaining the local density of the data point in the current layer as an example, the obtaining methods of the voltage layer and the temperature layer are the same.

[0055] Specifically, the aggregation degree satisfies the following relational expression:

[0056] ;

[0057] In the formula, represents the aggregation degree of the th clustering cluster, represents the distance between the th data in the th clustering cluster and the clustering center , represents the number of data in the th clustering cluster.

[0058] Taking any clustering cluster as the target clustering cluster, obtain the reciprocal of the distance between the clustering center of the target clustering cluster and the clustering center of the neighboring clustering cluster, and the reciprocal of the difference between the aggregation degree of the target clustering cluster and the aggregation degree of the neighboring clustering cluster;

[0059] Take the reciprocal of the distance between the clustering centers as the weight of the reciprocal of the difference between the aggregation degrees, and perform weighted summation as the local density of the clustering cluster.

[0060] Specifically, the local density satisfies the following relational expression:

[0061] ;

[0062] In the formula, represents the local density of the target clustering cluster, represents the clustering center of the th clustering cluster and the clustering center of the nearest neighboring clustering cluster of the th clustering cluster, represents the aggregation degree of the th clustering cluster, represents the aggregation degree of the nearest neighboring clustering cluster of the th clustering cluster, represents the total number of clustering clusters.

[0063] That is to say, the smaller the difference in the degree of aggregation between each cluster and its nearest neighbor cluster, the higher the local density of the data point. The closer the distance between the cluster center of the cluster and the cluster center of its nearest neighbor cluster, the greater the corresponding weight of the cluster. Exemplarily, taking the current layer data as an example, the higher the local density of the data point in the current layer, the smaller the value is required to better capture local anomalies. Accordingly, the initial value needs to be corrected.

[0064] S4: Correct the initial value according to the local density to obtain the corrected significant value. Use the significant value as the parameter value in the HWKS algorithm, obtain the anomaly degree of each level of data, perform weighted summation on the anomaly degrees of each level of data to obtain the comprehensive anomaly degree of each level of data, and judge whether the data point is abnormal data based on the comprehensive anomaly degree, and monitor the detection device.

[0065] Furthermore, in this embodiment, the HWKS algorithm is well-known to those skilled in the art, and the specific implementation steps will not be described in detail.

[0066] The significant value satisfies the following relational expression:

[0067] ;

[0068] In the formula, represents the significant value of each level of data, represents the preset hyperparameter, represents the local density of the target cluster, represents the initial value of each level of data, represents the exponential function with as the base.

[0069] That is to say, in this embodiment, , which can be adjusted according to the actual scenario.

[0070] The comprehensive anomaly degree satisfies the following relational expression:

[0071] ;

[0072] In the formula, represents the comprehensive anomaly degree, represents the anomaly degree of the voltage level, represents the anomaly degree of the current level, represents the anomaly degree of the temperature level, respectively represent the voltage level weight, the current level weight, and the temperature level weight.

[0073] That is to say, in this embodiment, .

[0074] In response to the comprehensive anomaly degree being greater than the anomaly threshold, it is abnormal data; otherwise, it is normal data. After obtaining the abnormal data, the warning device is triggered to perform real-time monitoring on the detection device.

[0075] In this embodiment, the anomaly threshold is 1, which can be adjusted according to those skilled in the art.

[0076] The present invention also provides a monitoring system for the operating state of a battery automatic short-circuit and reverse-polarity detection device. As Figure 2 shown, the system includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for monitoring the operating state of the battery automatic short-circuit and reverse-polarity detection device according to the first aspect of the present invention is implemented.

[0077] The system also includes other components well-known to those skilled in the art such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be elaborated here.

[0078] In the present invention, the aforementioned memory can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium can be a part of the device or accessible or connectable to the device. Any application or module described in the present invention can be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.

[0079] In the description of this specification, the meanings of "a plurality of" and "several" are at least two, such as two, three or more, etc., unless otherwise specifically defined.

[0080] Although this specification has shown and described multiple embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, alterations and alternative ways without departing from the spirit and concept of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in the practice of the present invention.

Claims

1. Method for monitoring the operating state of a battery automatic short-circuit reverse-polarity detection device, characterized in that, Including: Obtain the historical operation data of each level of the battery automatic short - circuit and reverse - polarity detection device, where the historical operation data of each level includes: current, voltage, and temperature; Based on the historical change rate and change frequency of the historical operation data at the level where it is located in the HWKS algorithm, it is used to calculate the fluctuation degree of the historical operation data at the level where it is located, and the fluctuation degree is used as the initial value; The initial value, including the steps of: Select the historical operation data in a preset time period before the data point acquisition moment in the historical operation data, draw a line chart for the selected historical operation data, obtain the extreme points in the line chart, and calculate the slope between adjacent extreme points to form a slope set; The square of the difference between the sum of the absolute values of the positive slopes and the sum of the absolute values of the negative slopes is taken as the slope difference, and the square of the difference in the probabilities of the occurrence of the positive slopes and the negative slopes is taken as the probability difference. The square root of the sum of the slope difference and the probability difference is used to evaluate the degree of fluctuation of each level of data, and this degree of fluctuation is taken as the initial value; Cluster the historical operation data of each level to obtain multiple clustering clusters, obtain the degree of aggregation between each clustering cluster, and calculate the local density of the clustering cluster according to the distance difference and aggregation degree difference between the clustering centers of each clustering cluster; The local density of the clustering cluster includes: Taking any clustering cluster as the target clustering cluster, obtain the reciprocal of the distance between the clustering center of the target clustering cluster and the clustering center of the neighboring clustering cluster, and the reciprocal of the difference between the aggregation degree of the target clustering cluster and the aggregation degree of the neighboring clustering cluster; Taking the reciprocal of the distance between the clustering centers as the weight of the reciprocal of the difference between the aggregation degrees, and performing weighted summation as the local density of the clustering cluster; Modify the initial value according to the local density to obtain the modified significant value. Use the significant value as the parameter value in the HWKS algorithm to obtain the abnormality degree of each level of data, perform weighted summation on the abnormality degrees of each level of data to be used to obtain the comprehensive abnormality degree of each level of data, determine whether the data point is abnormal data based on the comprehensive abnormality degree, and monitor the detection device.

2. The method for monitoring the operating state of the battery automatic short-circuit reverse-polarity detection device according to claim 1, characterized in that, The obtaining the degree of aggregation between each clustering cluster includes: Select the historical operation data in a preset time period before the data point acquisition moment in the historical operation data, cluster the selected local data to obtain several clustering clusters, take any clustering cluster as the target clustering cluster, and obtain the average value of the distances between each data point in the target clustering cluster and the clustering center of the target clustering cluster as the degree of aggregation.

3. The method for monitoring the operating state of the battery automatic short-circuit reverse-polarity detection device according to claim 1, characterized in that, The significant value satisfies the following relational expression: ； In the formula, represents the significance value of each level of data, represents a preset hyperparameter, represents the local density of the target clustering cluster, represents the initial value of each level of data, represents the exponential function with as the base.

4. The method for monitoring the operating state of the battery automatic short-circuit reverse-polarity detection device according to claim 1, characterized in that, The comprehensive abnormality degree satisfies the following relational expression: ； Wherein, represents the comprehensive abnormality degree, represents the abnormality degree of the voltage level, represents the abnormality degree of the current level, represents the abnormality degree of the temperature level, respectively represent the weight of the voltage level, the weight of the current level, and the weight of the temperature level.

5. The method for monitoring the operating state of the battery automatic short-circuit reverse-polarity detection device according to claim 1, wherein Based on the comprehensive abnormality degree, determine whether the data point is abnormal data, and monitor the detection device, including: In response to the comprehensive abnormality degree being greater than the abnormality threshold, it is abnormal data, otherwise, it is normal data; trigger the warning device after obtaining the abnormal data, and perform real - time monitoring on the detection device.

6. Battery automatic short - circuit reverse - polarity detection device operation status monitoring system, characterized in that, Including: A processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for monitoring the operation state of the battery automatic short - circuit and reverse - polarity detection device according to any one of claims 1 - 5 is implemented.

Images