A terminal and a method for generating a threshold of a station monitoring index

Abstract

The application discloses a kind of yard monitoring index threshold generation method and terminal, obtain the historical detection index data of yard in preset time period, according to time series anomaly detection algorithm identification abnormal point in historical detection index data and the abnormal interval of each abnormal point;According to the interval upper bound and interval lower bound of the abnormal interval of each abnormal point, the abnormal interval upper bound sequence and abnormal interval lower bound sequence corresponding to historical detection index data are obtained;According to abnormal interval upper bound value sequence, determine abnormal detection upper threshold, according to abnormal interval lower bound sequence, obtain abnormal detection lower threshold, obtain safety index threshold interval.The application corresponds to the judgment standard whether the current detection index data of the yard is abnormal, to assist accurately identifying the abnormal situation of yard monitoring index data, guarantee the safe operation of yard.

Description

Technical Field

[0001] This invention relates to the field of electronic information technology, and in particular to a method and terminal for generating thresholds for monitoring indicators at a site. Background Technology

[0002] With the increasing popularity of electric vehicles, the coverage of photovoltaic-energy storage-charging-testing stations, as one of the infrastructures for vehicle charging and testing, is expanding. Ensuring the safe operation of these stations is a crucial task in daily operation and maintenance. Therefore, real-time monitoring of numerous safety indicators at each station facilitates refined management and improves the overall safety management level of the stations.

[0003] However, due to the influence of personalized configurations, the situation of each station is different; moreover, the data distribution of station safety monitoring indicators itself has various distribution characteristics such as single-peak distribution and double-peak distribution, and the data fluctuation range is large, making it difficult to detect whether there are any abnormalities in the data changes in a timely manner during the monitoring process; how to accurately identify the abnormalities in the data of station monitoring indicators has become the key to ensuring the safe operation of the station. Summary of the Invention

[0004] The technical problem to be solved by this invention is to propose a method and terminal for generating thresholds for station monitoring indicators, which is crucial for accurately identifying abnormal data of station monitoring indicators and ensuring the safe operation of stations.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A method for generating thresholds for monitoring indicators at a facility includes the following steps:

[0007] S1. Obtain historical detection index data of the station within a preset time period, and identify abnormal points and abnormal intervals of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm.

[0008] S2. Based on the upper and lower bounds of the abnormal intervals of each abnormal point, obtain the upper and lower bound sequences of the abnormal intervals corresponding to the historical detection index data.

[0009] S3. Determine the upper bound threshold for anomaly detection based on the upper bound value sequence of the anomaly interval, and obtain the lower bound threshold for anomaly detection based on the lower bound sequence of the anomaly interval, thereby obtaining the safety index threshold range.

[0010] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0011] A terminal for generating threshold indicators for monitoring stations includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps:

[0012] S1. Obtain historical detection index data of the station within a preset time period, and identify abnormal points and abnormal intervals of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm.

[0013] S2. Based on the upper and lower bounds of the abnormal intervals of each abnormal point, obtain the upper and lower bound sequences of the abnormal intervals corresponding to the historical detection index data.

[0014] S3. Determine the upper bound threshold for anomaly detection based on the upper bound value sequence of the anomaly interval, and obtain the lower bound threshold for anomaly detection based on the lower bound sequence of the anomaly interval, thereby obtaining the safety index threshold range.

[0015] The beneficial effects of this invention are as follows: It proposes a method and terminal for generating threshold values ​​for monitoring indicators at a facility. The method uses a time series algorithm to identify abnormal points and abnormal intervals of historical monitoring indicator data of the facility within a preset time period. Then, by integrating the upper and lower bound sequences of the abnormal intervals, the upper and lower threshold values ​​of the safety indicator threshold interval are determined. This method provides a criterion for judging whether the current monitoring indicator data of the corresponding facility is abnormal, thereby assisting in accurately identifying abnormal situations in the monitoring indicator data of the facility and ensuring the safe operation of the facility. Attached Figure Description

[0016] Figure 1 This is a schematic diagram illustrating the steps of a method for generating threshold values ​​for station monitoring indicators according to the present invention.

[0017] Figure 2 This is a system block diagram of a station monitoring index threshold generation terminal according to the present invention.

[0018] Label Explanation:

[0019] 1. A terminal for generating threshold values ​​for monitoring indicators at a field station; 2. A memory; 3. A processor. Detailed Implementation

[0020] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0021] Please refer to Figure 1 A method for generating thresholds for station monitoring indicators, comprising the following steps:

[0022] S1. Obtain historical detection index data of the station within a preset time period, and identify abnormal points and abnormal intervals of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm.

[0023] S2. Based on the upper and lower bounds of the abnormal intervals of each abnormal point, obtain the upper and lower bound sequences of the abnormal intervals corresponding to the historical detection index data.

[0024] S3. Determine the upper bound threshold for anomaly detection based on the upper bound value sequence of the anomaly interval, and obtain the lower bound threshold for anomaly detection based on the lower bound sequence of the anomaly interval, thereby obtaining the safety index threshold range.

[0025] As can be seen from the above description, the beneficial effects of the present invention are as follows: It proposes a method for generating threshold values ​​for monitoring indicators at a facility. This method utilizes a time series algorithm to identify anomalies in the historical monitoring indicator data of the facility within a preset time period and the anomaly intervals of each anomaly. It uses the changing intervals of multiple anomalies that occur as the facility's operating time increases to understand the abnormal changes in the monitoring indicator data. Then, by integrating the upper and lower bound sequences of the anomaly intervals, it determines the upper and lower threshold values ​​of the safety indicator threshold interval. That is, the safety indicator threshold interval represents the range where the data size concentration occurs after different anomalies. This serves as a criterion for judging whether the current monitoring indicator data of the corresponding facility is abnormal, thereby assisting in accurately identifying anomalies in the monitoring indicator data of the facility and ensuring the safe operation of the facility.

[0026] Furthermore, step S3 specifically includes:

[0027] The maximum value in the sequence of upper bound values ​​of the abnormal interval is selected as the upper bound threshold for anomaly detection, and the minimum value in the sequence of lower bound values ​​of the abnormal interval is selected as the lower bound threshold for anomaly detection.

[0028] As can be seen from the above description, when determining the upper and lower bounds, the safety index threshold range is generated by referencing the extreme values ​​corresponding to the upper and lower bound sequences of the abnormal interval. This improves the coverage of abnormal situations and helps to more reliably ensure the safe operation of the station.

[0029] Further, step S2 specifically includes:

[0030] S21. Remove the maximum value from the upper bound of the abnormal interval of all the abnormal points to obtain the upper bound sequence of the abnormal interval;

[0031] S22. Remove the minimum value from the lower bound of the abnormal intervals of all the abnormal points to obtain the sequence of lower bounds of the abnormal intervals.

[0032] As can be seen from the above description, by eliminating extreme values, the rationality of the selected data can be improved, and the final judgment can be avoided due to a few special cases, thereby improving the accuracy and applicability of the safety indicator threshold range.

[0033] Furthermore, it also includes:

[0034] S4. Obtain the current detection index data of the station, detect whether there is any data anomaly in the current detection index data according to the safety index threshold range, if not, add the current detection index data to the historical detection index data, identify the abnormal points in the historical detection index data and the abnormal range of each abnormal point according to the time series anomaly detection algorithm, and execute steps S2 and S3.

[0035] As described above, the generated safety indicator threshold range is used to detect whether there are any anomalies in the current detection indicator data. If the value does not fall within the safety indicator threshold range, it indicates that the current data of the site is normal. At the same time, the current detection indicator data can also be used as the latest historical detection indicator data to update the safety indicator threshold range and ensure the reliability of subsequent anomaly detection.

[0036] Furthermore, the procedure prior to step S1 includes:

[0037] S0. Acquire the detection index data of the site within a preset period and accumulate the number of acquisitions. Adjust the duration of the preset time period based on the number of acquisitions.

[0038] As can be seen from the above description, as the amount of detection index data at the site increases, the duration of the preset time period is adaptively adjusted, that is, the historical detection index data referenced is increased, so as to obtain the changes in the detection index data more comprehensively and maintain the accuracy of the safety index threshold range.

[0039] Please refer to Figure 2 A terminal 1 for generating threshold indicators for monitoring stations includes a memory 2, a processor 3, and a computer program stored in the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, it performs the following steps:

[0040] S1. Obtain historical detection index data of the station within a preset time period, and identify abnormal points and abnormal intervals of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm.

[0041] S2. Based on the upper and lower bounds of the abnormal intervals of each abnormal point, obtain the upper and lower bound sequences of the abnormal intervals corresponding to the historical detection index data.

[0042] S3. Determine the upper bound threshold for anomaly detection based on the upper bound value sequence of the anomaly interval, and obtain the lower bound threshold for anomaly detection based on the lower bound sequence of the anomaly interval, thereby obtaining the safety index threshold range.

[0043] As can be seen from the above description, the beneficial effects of the present invention are as follows: It utilizes time series algorithms to identify anomalies in the historical monitoring indicator data of the facility within a preset time period and the anomaly intervals of each anomaly. It uses the changing intervals of multiple anomalies that occur as the facility's operating time increases to understand the abnormal changes in the facility's monitoring indicator data. Then, by integrating the upper and lower bound sequences of the anomaly intervals, it determines the upper and lower thresholds of the safety indicator threshold interval. That is, the safety indicator threshold interval represents the range where the data size is concentrated after different anomalies occur. This serves as the criterion for judging whether the current monitoring indicator data of the corresponding facility is abnormal, thereby assisting in accurately identifying anomalies in the facility's monitoring indicator data and ensuring the safe operation of the facility.

[0044] Furthermore, step S3 specifically includes:

[0045] The maximum value in the sequence of upper bound values ​​of the abnormal interval is selected as the upper bound threshold for anomaly detection, and the minimum value in the sequence of lower bound values ​​of the abnormal interval is selected as the lower bound threshold for anomaly detection.

[0046] As can be seen from the above description, when determining the upper and lower bounds, the safety index threshold range is generated by referencing the extreme values ​​corresponding to the upper and lower bound sequences of the abnormal interval. This improves the coverage of abnormal situations and helps to more reliably ensure the safe operation of the station.

[0047] Further, step S2 specifically includes:

[0048] S21. Remove the maximum value from the upper bound of the abnormal interval of all the abnormal points to obtain the upper bound sequence of the abnormal interval;

[0049] S22. Remove the minimum value from the lower bound of the abnormal intervals of all the abnormal points to obtain the sequence of lower bounds of the abnormal intervals.

[0050] As can be seen from the above description, by eliminating extreme values, the rationality of the selected data can be improved, and the final judgment can be avoided due to a few special cases, thereby improving the accuracy and applicability of the safety indicator threshold range.

[0051] Furthermore, it also includes:

[0052] S4. Obtain the current detection index data of the station, detect whether there is any data anomaly in the current detection index data according to the safety index threshold range, if not, add the current detection index data to the historical detection index data, identify the abnormal points in the historical detection index data and the abnormal range of each abnormal point according to the time series anomaly detection algorithm, and execute steps S2 and S3.

[0053] As described above, the generated safety indicator threshold range is used to detect whether there are any anomalies in the current detection indicator data. If the value does not fall within the safety indicator threshold range, it indicates that the current data of the site is normal. At the same time, the current detection indicator data can also be used as the latest historical detection indicator data to update the safety indicator threshold range and ensure the reliability of subsequent anomaly detection.

[0054] Furthermore, the procedure prior to step S1 includes:

[0055] S0. Acquire the detection index data of the site within a preset period and accumulate the number of acquisitions. Adjust the duration of the preset time period based on the number of acquisitions.

[0056] As can be seen from the above description, as the amount of detection index data at the site increases, the duration of the preset time period is adaptively adjusted, that is, the historical detection index data referenced is increased, so as to obtain the changes in the detection index data more comprehensively and maintain the accuracy of the safety index threshold range.

[0057] Please refer to Figure 1 Embodiment 1 of the present invention is as follows:

[0058] A method for generating thresholds for monitoring indicators at a facility includes the following steps:

[0059] S0. While acquiring the station's detection index data within a preset period, accumulate the number of acquisitions, and adjust the duration of the preset time period based on the number of acquisitions.

[0060] In this embodiment, the length of the preset cycle can be determined based on which data point in the selected site's testing index data, such as the site's PCS insulation value; the operating characteristics of the corresponding equipment will also affect the cycle length. Furthermore, the method of this embodiment is not only applicable to site testing index data, but also to monitoring scenarios involving data such as battery main positive and main negative insulation values, maximum temperature of individual battery cells, battery compartment temperature, equipment compartment temperature, battery compartment humidity, equipment compartment humidity, hydrogen concentration, and carbon monoxide solubility.

[0061] S1. Obtain historical detection index data of the station within a preset time period, and identify abnormal points and abnormal intervals of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm.

[0062] In this embodiment, the preset time can be a period from a past time point to the current time, or it can be a time period where some data are undergoing different changes, selected as the preset time based on data changes. As for the time series anomaly detection algorithm, the preferred algorithm is the SH-ESD algorithm. SH-ESD is an anomaly detection method, short for "Seasonal Hybrid Extreme Studentized Deviate". This method is used to detect outliers in time series data, especially for datasets with seasonality. Its specific application process can be found in the journal article titled "Automatic Anomaly Detection in the Cloud Via Statistical Learning," authored by Jordan Hochenbaum, Owen S., Vallis Arun, Kejariwal, and Twitter Inc.

[0063] Based on the SH-ESD algorithm, the process for determining outliers and the outlier intervals for each outlier in historical detection index data is as follows:

[0064] First, define the median absolute difference (MAD) in historical monitoring indicator data using MAD:

[0065] MAD = median(|X i-median(X));

[0066] Where Xi is the i-th data item in the historical detection index data;

[0067] Then, the residual between each data point in the historical test data and the median absolute value is calculated. If the absolute value of the residual of a certain data point is greater than the preset value, it indicates that the data point is abnormal.

[0068] Once an outlier is identified, the absolute value of the residual of that outlier can be used as the endpoint of the interval for iteration. That is, the outlier interval corresponding to the second outlier is the interval formed by the absolute value of the residual of the first outlier and the absolute value of its own residual as the upper and lower bounds, and so on. Therefore, when determining the safety index threshold interval, in addition to using the operation methods of steps S2 and S3, the upper and lower bounds of the outlier interval corresponding to the last outlier can also be selected as the endpoints of the safety index threshold interval.

[0069] S2. Based on the upper and lower bounds of the abnormal interval for each abnormal point, obtain the upper and lower bound sequences of the abnormal intervals for the corresponding historical detection index data.

[0070] In this embodiment, the maximum value in the upper bound sequence of the abnormal interval is selected as the upper bound threshold for anomaly detection, and the minimum value in the lower bound sequence of the abnormal interval is selected as the lower bound threshold for anomaly detection.

[0071] S3. Determine the upper bound threshold for anomaly detection based on the upper bound value sequence of the anomaly interval, and obtain the lower bound threshold for anomaly detection based on the lower bound sequence of the anomaly interval, thus obtaining the safety indicator threshold range.

[0072] In this embodiment, as a preferred design, the maximum value in the sequence of upper bound values ​​of the abnormal interval is selected as the upper bound threshold for anomaly detection, and the minimum value in the sequence of lower bound values ​​of the abnormal interval is selected as the lower bound threshold for anomaly detection.

[0073] S4. Obtain the current detection index data of the station, and check whether there is any data anomaly in the current detection index data according to the safety index threshold range. If not, add the current detection index data to the historical detection index data, identify the abnormal points and the abnormal range of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm, and execute steps S2 and S3.

[0074] In this embodiment, the upper and lower bounds of the safety indicator threshold range are the upper and lower bounds of the site safety monitoring indicators; if the current detection indicator data is within the safety indicator threshold range, it indicates that the data is abnormal and an alarm maintenance should be triggered immediately.

[0075] Furthermore, in the above steps, the average and maximum values ​​in the upper bound sequence of the abnormal interval can be combined to form an upper warning interval, while the average and minimum values ​​in the lower bound sequence of the abnormal interval can be combined to form a lower warning interval. When the current detection index data does not fall within the upper or lower warning interval, although no alarm is triggered, a reminder will be sent to the operation and maintenance personnel. This allows the operation and maintenance personnel to combine other data related to the current detection index data to determine whether it is reasonable for the data to fall within the upper or lower warning interval, thereby predicting possible abnormal situations.

[0076] Please refer to Figure 2 Embodiment two of the present invention is as follows:

[0077] A terminal 1 for generating thresholds for monitoring indicators at a facility includes a memory 2, a processor 3, and a computer program stored in the memory 2 and capable of running on the processor 3. When the processor 3 executes the computer program, it implements a method for generating thresholds for monitoring indicators at a facility according to Embodiment 1.

[0078] In summary, the present invention provides a method and terminal for generating threshold values ​​for monitoring indicators at a facility. It utilizes a time series algorithm to identify anomalies in historical monitoring indicator data of the facility within a preset time period, and the anomaly intervals of each anomaly. By utilizing the changing intervals of multiple anomalies that occur over time, the method grasps the abnormal changes in the monitoring indicator data of the facility. Then, by integrating the upper and lower bound sequences of the anomaly intervals, it determines the upper and lower threshold values ​​of the safety indicator threshold interval. In other words, the safety indicator threshold interval represents the range of data size concentrations after different anomalies occur. This serves as a criterion for judging whether the current monitoring indicator data of the corresponding facility is abnormal, thereby assisting in accurately identifying anomalies in the monitoring indicator data of the facility and ensuring the safe operation of the facility.

[0079] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for generating threshold values ​​for monitoring indicators at a facility, characterized in that, Includes the following steps: S1. Obtain historical detection index data of the station within a preset time period, and identify abnormal points and abnormal intervals of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm. S2. Based on the upper and lower bounds of the abnormal intervals of each abnormal point, obtain the upper and lower bound sequences of the abnormal intervals corresponding to the historical detection index data. Step S2 specifically includes: S21. Remove the maximum value from the upper bound of the abnormal interval of all the abnormal points to obtain the upper bound sequence of the abnormal interval; S22. Remove the minimum value from the lower bound of the abnormal intervals of all the abnormal points to obtain the sequence of lower bounds of the abnormal intervals; S3. Determine the upper bound threshold for anomaly detection based on the upper bound sequence of the anomaly interval, and obtain the lower bound threshold for anomaly detection based on the lower bound sequence of the anomaly interval, thereby obtaining the safety index threshold range. Step S3 specifically includes: The maximum value in the upper bound sequence of the abnormal interval is selected as the upper bound threshold for anomaly detection, and the minimum value in the lower bound sequence of the abnormal interval is selected as the lower bound threshold for anomaly detection. The average and maximum values ​​in the upper bound sequence of the abnormal interval are combined to form an upper bound warning interval, and the average and minimum values ​​in the lower bound sequence of the abnormal interval are combined to form a lower bound warning interval.

2. The method for generating threshold values ​​for station monitoring indicators according to claim 1, characterized in that, Also includes: S4. Obtain the current detection index data of the station, detect whether there is any data anomaly in the current detection index data according to the safety index threshold range, if not, add the current detection index data to the historical detection index data, identify the abnormal points in the historical detection index data and the abnormal range of each abnormal point according to the time series anomaly detection algorithm, and execute steps S2 and S3.

3. The method for generating threshold values ​​for station monitoring indicators according to claim 1, characterized in that, The procedure preceding step S1 also includes: S0. Acquire the detection index data of the site within a preset period and accumulate the number of acquisitions. Adjust the duration of the preset time period based on the number of acquisitions.

4. A terminal for generating thresholds for monitoring indicators at a field station, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it performs the following steps: S1. Obtain historical detection index data of the station within a preset time period, and identify abnormal points and abnormal intervals of each abnormal point in the historical detection index data according to the time series anomaly detection algorithm. S2. Based on the upper and lower bounds of the abnormal intervals of each abnormal point, obtain the upper and lower bound sequences of the abnormal intervals corresponding to the historical detection index data. Step S2 specifically includes: S21. Remove the maximum value from the upper bound of the abnormal interval of all the abnormal points to obtain the upper bound sequence of the abnormal interval; S22. Remove the minimum value from the lower bound of the abnormal intervals of all the abnormal points to obtain the sequence of lower bounds of the abnormal intervals; S3. Determine the upper bound threshold for anomaly detection based on the upper bound sequence of the anomaly interval, and obtain the lower bound threshold for anomaly detection based on the lower bound sequence of the anomaly interval, thereby obtaining the safety index threshold range. Step S3 specifically includes: The maximum value in the upper bound sequence of the abnormal interval is selected as the upper bound threshold for anomaly detection, and the minimum value in the lower bound sequence of the abnormal interval is selected as the lower bound threshold for anomaly detection. The average and maximum values ​​in the upper bound sequence of the abnormal interval are combined to form an upper bound warning interval, and the average and minimum values ​​in the lower bound sequence of the abnormal interval are combined to form a lower bound warning interval.

5. A terminal for generating threshold indicators for station monitoring according to claim 4, characterized in that, Also includes: S4. Obtain the current detection index data of the station, detect whether there is any data anomaly in the current detection index data according to the safety index threshold range, if not, add the current detection index data to the historical detection index data, identify the abnormal points in the historical detection index data and the abnormal range of each abnormal point according to the time series anomaly detection algorithm, and execute steps S2 and S3.

6. A terminal for generating threshold indicators for station monitoring according to claim 4, characterized in that, The procedure preceding step S1 also includes: S0. Acquire the detection index data of the site within a preset period and accumulate the number of acquisitions. Adjust the duration of the preset time period based on the number of acquisitions.

Images