Operation and maintenance alarm method and apparatus, and storage medium and electronic device

Abstract

An operation and maintenance alarm method and apparatus, and a storage medium and an electronic device. The method comprises: acquiring an operation and maintenance data set to be processed, and by means of a target algorithm, performing data cleaning processing on operation and maintenance data in the operation and maintenance data set to be processed, so as to obtain a processed operation and maintenance data set (S101); performing a stability test on a time sequence of the operation and maintenance data in the processed operation and maintenance data set, so as to obtain a test result, and determining a first operation and maintenance data set from the processed operation and maintenance data set on the basis of the test result (S102); performing feature extraction processing on the first operation and maintenance data set by means of a principal component analysis method, so as to obtain a plurality of operation and maintenance data features, and performing feature selection processing on the plurality of operation and maintenance data features by means of a K-means clustering algorithm, so as to obtain a target operation and maintenance data set (S103); and performing alarm analysis on the basis of the target operation and maintenance data set, so as to obtain an alarm analysis result (S104). The technical problem in the prior art of the accuracy of an operation and maintenance alarm being relatively low due to the fact that the accuracy of data pre-processing performed on the basis of a fixed rule is low is solved.

Description

Operation and maintenance alarm method, device, storage medium and electronic equipment

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on December 14, 2023, with application number 2023117238389, and application name “Operation and maintenance alarm method, device, storage medium and electronic equipment”, all contents of which are incorporated by reference in this application. Technical Field

[0002] The present application relates to the field of data processing technology, and in particular to an operation and maintenance alarm method, device, storage medium and electronic equipment. Background Art

[0003] Data preprocessing is the process of detecting and correcting dirty data and is the foundation of data analysis and management. Currently, existing technologies typically perform data preprocessing according to fixed rules during production operations, and use this preprocessed data for operations alarm analysis. This results in low data preprocessing accuracy, which in turn leads to low accuracy in operations alarms and a high false alarm rate.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] The embodiments of the present application provide an operation and maintenance alarm method, device, storage medium and electronic device to at least solve the technical problem in the prior art of performing data preprocessing based on fixed rules, resulting in low accuracy of data preprocessing and resulting in low accuracy of operation and maintenance alarms.

[0006] According to one aspect of an embodiment of the present application, an operation and maintenance alarm method is provided, including: obtaining an operation and maintenance data set to be processed, and performing data cleaning processing on the operation and maintenance data in the operation and maintenance data set to be processed by a target algorithm to obtain a processed operation and maintenance data set, wherein the target algorithm at least includes a quadratic interpolation algorithm; performing a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, and determining a first operation and maintenance data set from the processed operation and maintenance data set based on the test result, wherein the test result is used to characterize whether the time series is a stationary time series; performing feature extraction processing on the first operation and maintenance data set by a principal component analysis method to obtain multiple operation and maintenance data features, and performing feature selection processing on the multiple operation and maintenance data features by a K-means clustering algorithm to obtain a target operation and maintenance data set; performing alarm analysis based on the target operation and maintenance data set to obtain an alarm analysis result, wherein the alarm analysis result is used to indicate whether an alarm is issued.

[0007] In some embodiments, data cleaning processing is performed on the operation and maintenance data in the operation and maintenance data set to be processed by a target algorithm to obtain a processed operation and maintenance data set, including: calculating the missing rate of each operation and maintenance data in the operation and maintenance data set to be processed; based on the missing rate of each operation and maintenance data, eliminating the operation and maintenance data with a missing rate greater than a missing rate threshold from the operation and maintenance data set to be processed to obtain a second operation and maintenance data set; and filling missing values ​​in the operation and maintenance data in the second operation and maintenance data set by a quadratic interpolation algorithm to obtain a processed operation and maintenance data set.

[0008] In some embodiments, the operation and maintenance data in the second operation and maintenance data set is filled with missing values ​​using a quadratic interpolation algorithm to obtain a processed operation and maintenance data set, including: determining the missing values ​​of the operation and maintenance data in the second operation and maintenance data set, and determining the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment adjacent to the missing values; calculating the mean of the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment, and filling the missing values ​​based on the mean to obtain the processed operation and maintenance data set.

[0009] In some embodiments, a stationarity test is performed on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, including: calculating the probability value of the time series having a unit root through a unit root test algorithm, and comparing the probability value with a preset threshold; if the probability value is less than the preset threshold, then the time series is not a stationary time series as a test result; if the probability value is greater than or equal to the preset threshold, then the time series is a stationary time series as a test result.

[0010] In some embodiments, based on the detection results, a first operation and maintenance data set is determined from the processed operation and maintenance data set, including: based on the detection results, determining target operation and maintenance data from the processed operation and maintenance data set, and eliminating the target operation and maintenance data to obtain eliminated operation and maintenance data, wherein the target operation and maintenance data is the operation and maintenance data corresponding to a time series that is not a stationary time series; and forming the first operation and maintenance data set based on the eliminated operation and maintenance data.

[0011] In some embodiments, feature extraction processing is performed on the first operation and maintenance data set through principal component analysis to obtain multiple operation and maintenance data features, including: data dimensionality reduction processing is performed on the operation and maintenance data in the first operation and maintenance data set through principal component analysis to obtain a reduced-dimensional operation and maintenance data set; feature extraction processing is performed on the reduced-dimensional operation and maintenance data set to obtain multiple operation and maintenance data features.

[0012] In some embodiments, feature selection processing is performed on multiple operation and maintenance data features through the K-means clustering algorithm to obtain a target operation and maintenance data set, including: determining the initial number of clusters N based on the CH indicator, and randomly generating N initial cluster centers through the K-means clustering algorithm, where N is a positive integer; calculating the distance between each operation and maintenance data feature and the N initial cluster centers, and determining the cluster corresponding to each operation and maintenance data feature based on the distance; iterative clustering is performed based on the cluster clusters until a preset iteration termination condition is met to obtain the target operation and maintenance data set.

[0013] According to another aspect of an embodiment of the present application, an operation and maintenance alarm device is also provided, including: an acquisition module, used to acquire an operation and maintenance data set to be processed, and perform data cleaning processing on the operation and maintenance data in the operation and maintenance data set to be processed through a target algorithm to obtain a processed operation and maintenance data set, wherein the target algorithm at least includes a quadratic interpolation algorithm; a determination module, used to perform stationarity detection on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a detection result, and determine a first operation and maintenance data set from the processed operation and maintenance data set based on the detection result, wherein the detection result is used to characterize whether the time series is a stationary time series; a first processing module, used to perform feature extraction processing on the first operation and maintenance data set through a principal component analysis method to obtain multiple operation and maintenance data features, and perform feature selection processing on the multiple operation and maintenance data features through a K-means clustering algorithm to obtain a target operation and maintenance data set; a second processing module, used to perform alarm analysis based on the target operation and maintenance data set to obtain an alarm analysis result, wherein the alarm analysis result is used to indicate whether an alarm is issued.

[0014] According to another aspect of an embodiment of the present application, a computer-readable storage medium is further provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned operation and maintenance alarm method when running.

[0015] According to another aspect of an embodiment of the present application, an electronic device is also provided, which includes one or more processors; a memory for storing one or more programs, which, when the one or more programs are executed by one or more processors, enables the one or more processors to run the programs, wherein the programs are configured to execute the above-mentioned operation and maintenance alarm method during operation.

[0016] In an embodiment of the present application, a method of data preprocessing based on quadratic interpolation and elimination of missing rate is adopted. First, an operation and maintenance data set to be processed is obtained, and the operation and maintenance data in the operation and maintenance data set to be processed is cleaned by a target algorithm to obtain a processed operation and maintenance data set. Then, a stationarity test is performed on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result. Based on the test result, a first operation and maintenance data set is determined from the processed operation and maintenance data set. Then, a feature extraction process is performed on the first operation and maintenance data set by the principal component analysis method to obtain multiple operation and maintenance data features. Then, a feature selection process is performed on multiple operation and maintenance data features by the K-means clustering algorithm to obtain a target operation and maintenance data set. Then, an alarm analysis is performed based on the target operation and maintenance data set to obtain an alarm analysis result. Among them, the target algorithm includes at least a quadratic interpolation algorithm, the test result is used to characterize whether the time series is a stationary time series, and the alarm analysis result is used to indicate whether an alarm is issued.

[0017] In the above process, the operation and maintenance data in the operation and maintenance data set to be processed are cleaned by the quadratic interpolation algorithm, and the time series of the operation and maintenance data in the processed operation and maintenance data set are tested for stationarity, which can effectively improve the accuracy of data preprocessing, reduce the interference of abnormal data on subsequent alarm analysis, and reduce the errors caused by abnormal data. Then, the data set that has undergone data cleaning and stationarity testing is subjected to feature generation and selection, so that alarm analysis can be performed based on the target operation and maintenance data set, effectively improving the accuracy of operation and maintenance alarms and reducing the false alarm rate of operation and maintenance alarms.

[0018] It can be seen that the technical solution of the present application has achieved the goal of reducing the false alarm rate of operation and maintenance alarms, thereby achieving the technical effect of improving the accuracy of operation and maintenance alarms, and further solved the technical problem in the prior art of performing data preprocessing based on fixed rules, resulting in low accuracy of data preprocessing and resulting in low accuracy of operation and maintenance alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] FIG1 is a flow chart of an optional operation and maintenance alarm method according to an embodiment of the present application;

[0021] FIG2 is a schematic diagram of an optional data preprocessing process according to an embodiment of the present application;

[0022] FIG3 is a schematic diagram of an optional operation and maintenance alarm device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set up between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0026] Example 1

[0027] According to an embodiment of the present application, an embodiment of an operation and maintenance alarm method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] FIG1 is a flow chart of an optional operation and maintenance alarm method according to an embodiment of the present application. As shown in FIG1 , the method includes the following steps:

[0029] Step S101: obtaining an operation and maintenance data set to be processed, and performing data cleaning processing on the operation and maintenance data in the operation and maintenance data set to be processed by a target algorithm to obtain a processed operation and maintenance data set, wherein the target algorithm at least includes a quadratic interpolation algorithm.

[0030] In the above steps, the operation and maintenance data set to be processed can be obtained through devices such as application systems, processors, and electronic devices. In some embodiments, the operation and maintenance data set to be processed is obtained through an operation and maintenance alarm system, where the operation and maintenance data can be operation and maintenance data of a target business, such as a payment business. Optionally, the operation and maintenance data in the operation and maintenance data set to be processed is cleaned using a target algorithm to obtain a processed operation and maintenance data set, where the target algorithm includes a quadratic interpolation algorithm and a method for eliminating missing rates. For example, a quadratic interpolation algorithm is used to clean continuous and slightly fluctuating data, and operation and maintenance data with a missing rate of more than 30% is eliminated. For operation and maintenance data with a missing rate of less than 30%, the average of two adjacent moments is used as a replacement.

[0031] Step S102: Perform a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, and determine a first operation and maintenance data set from the processed operation and maintenance data set based on the test result, wherein the test result is used to indicate whether the time series is a stationary time series.

[0032] In some embodiments, a first operation and maintenance data set can be determined based on the detection results obtained by performing a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set, wherein the first operation and maintenance data set is composed of the operation and maintenance data corresponding to the stationary time series. For example, by performing a stationarity test, the operation and maintenance data corresponding to the unstable time series can be eliminated to obtain the first operation and maintenance data set.

[0033] Step S103 , performing feature extraction processing on the first operation and maintenance data set by principal component analysis to obtain multiple operation and maintenance data features, and performing feature selection processing on the multiple operation and maintenance data features by K-means clustering algorithm to obtain a target operation and maintenance data set.

[0034] Step S104: performing alarm analysis based on the target operation and maintenance data set to obtain an alarm analysis result, wherein the alarm analysis result is used to indicate whether to issue an alarm.

[0035] In some embodiments, features are generated and selected for a data set that has undergone data cleaning and stationarity testing, so that alarm analysis can be performed based on a target operation and maintenance data set. For example, feature extraction processing is performed on the first operation and maintenance data set through principal component analysis to obtain multiple operation and maintenance data features. Feature selection processing is then performed on the multiple operation and maintenance data features through the K-means clustering algorithm to obtain a data set after data preprocessing (i.e., the target operation and maintenance data set), and alarm analysis is performed to determine whether to issue an alarm. For example, if the alarm analysis result is an alarm, a corresponding alarm prompt message is generated.

[0036] Based on the scheme defined by the above steps S101 to S104, it can be known that in the embodiment of the present application, a method of data preprocessing based on quadratic interpolation and elimination of missing rate is adopted, firstly, the operation and maintenance data set to be processed is obtained, and the operation and maintenance data in the operation and maintenance data set to be processed is cleaned by the target algorithm to obtain a processed operation and maintenance data set, and then the time series of the operation and maintenance data in the processed operation and maintenance data set is tested for stationarity to obtain a test result, and based on the test result, a first operation and maintenance data set is determined from the processed operation and maintenance data set, and then feature extraction processing is performed on the first operation and maintenance data set by the principal component analysis method to obtain multiple operation and maintenance data features, and feature selection processing is performed on multiple operation and maintenance data features by the K-means clustering algorithm to obtain a target operation and maintenance data set, and then an alarm analysis is performed based on the target operation and maintenance data set to obtain an alarm analysis result. Among them, the target algorithm includes at least a quadratic interpolation algorithm, the test result is used to characterize whether the time series is a stationary time series, and the alarm analysis result is used to indicate whether an alarm is issued.

[0037] It is easy to notice that in the above process, the operation and maintenance data in the operation and maintenance data set to be processed are cleaned by the quadratic interpolation algorithm, and the time series of the operation and maintenance data in the processed operation and maintenance data set are tested for stationarity. This can effectively improve the accuracy of data preprocessing, reduce the interference of abnormal data on subsequent alarm analysis, and reduce the errors caused by abnormal data. Then, the data set that has undergone data cleaning and stationarity testing is used for feature generation and selection, so that alarm analysis can be performed based on the target operation and maintenance data set, effectively improving the accuracy of operation and maintenance alarms and reducing the false alarm rate of operation and maintenance alarms.

[0038] It can be seen that the technical solution of the present application has achieved the goal of reducing the false alarm rate of operation and maintenance alarms, thereby achieving the technical effect of improving the accuracy of operation and maintenance alarms, and further solved the technical problem in the prior art of performing data preprocessing based on fixed rules, resulting in low accuracy of data preprocessing and resulting in low accuracy of operation and maintenance alarms.

[0039] In an optional embodiment, the operation and maintenance data in the operation and maintenance data set to be processed are cleaned by a target algorithm to obtain a processed operation and maintenance data set, including: calculating the missing rate of each operation and maintenance data in the operation and maintenance data set to be processed; based on the missing rate of each operation and maintenance data, eliminating the operation and maintenance data with a missing rate greater than a missing rate threshold from the operation and maintenance data set to be processed to obtain a second operation and maintenance data set; and filling the missing values ​​of the operation and maintenance data in the second operation and maintenance data set by a quadratic interpolation algorithm to obtain a processed operation and maintenance data set.

[0040] In an optional embodiment, the operation and maintenance data in the second operation and maintenance data set is filled with missing values ​​using a quadratic interpolation algorithm to obtain a processed operation and maintenance data set, including: determining the missing values ​​of the operation and maintenance data in the second operation and maintenance data set, and determining the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment adjacent to the missing values; calculating the mean of the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment, and filling the missing values ​​based on the mean to obtain the processed operation and maintenance data set.

[0041] In some embodiments, in the process of performing data cleaning processing on the operation and maintenance data in the operation and maintenance data set to be processed by a target algorithm to obtain a processed operation and maintenance data set, the missing rate of each operation and maintenance data in the operation and maintenance data set to be processed is first calculated, and then based on the missing rate of each operation and maintenance data, the operation and maintenance data with a missing rate greater than a missing rate threshold (for example, 30%) is eliminated from the operation and maintenance data set to be processed to obtain a second operation and maintenance data set composed of the eliminated operation and maintenance data, for example, the operation and maintenance data with a missing rate above 30% is eliminated to obtain the second operation and maintenance data set.

[0042] In some embodiments, the operation and maintenance data in the second operation and maintenance data set is filled with missing values ​​using a quadratic interpolation algorithm to obtain a processed operation and maintenance data set. Optionally, the missing values ​​of the operation and maintenance data in the second operation and maintenance data set are first determined, and the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment adjacent to the missing values ​​are determined. Then, the mean of the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment is calculated, and the missing values ​​are filled based on the mean to obtain the processed operation and maintenance data set. For example, after eliminating the operation and maintenance data with a missing rate of more than 30%, the mean of the two adjacent moments is used to replace the operation and maintenance data with a missing rate of less than 30%.

[0043] It should be noted that the data processed by the above process is evenly spaced, more consistent with the time series processing of the neural network model, and more realistically restores missing data, effectively improving the accuracy of data preprocessing, reducing the interference of abnormal data on subsequent alarm analysis, and reducing errors caused by abnormal data.

[0044] In an optional embodiment, a stationarity test is performed on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, including: calculating the probability value of the time series having a unit root through a unit root test algorithm, and comparing the probability value with a preset threshold; if the probability value is less than the preset threshold, then the time series is not a stationary time series as a test result; if the probability value is greater than or equal to the preset threshold, then the time series is a stationary time series as a test result.

[0045] In some embodiments, in the process of performing a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain the test results, the probability value of the time series having a unit root is first calculated by a unit root test algorithm, and the probability value is compared with a preset threshold (for example, 0.05). If the probability value is less than the preset threshold, the time series is not a stationary time series as a test result; if the probability value is greater than or equal to the preset threshold, the time series is a stationary time series as a test result.

[0046] In some embodiments, stability refers to the invariance of the statistical properties of the time series with respect to time translation. Unstable time series will cause problems such as "pseudo-regression" in regression analysis and other problems. Therefore, for the operation and maintenance data that have been preliminarily screened (i.e., the operation and maintenance data in the processed operation and maintenance data set), it is necessary to determine whether the time series of these data is stable. Optionally, a unit root test is performed with a confidence level of 95%, that is, when the P value is less than 0.05, the time series does not have a unit root, and the corresponding time series is unstable and will not participate in the subsequent data analysis. For example, if one point is taken every 5 minutes, then there are 12 points in 1 hour, and 12*24*30=8640 points if a month is 30 days. Then, the 5% of the traffic with the highest values ​​is removed, and the remaining 95% is in the normal traffic range, that is, the number of collected points is 8208 points, and 432 points belong to the abnormal traffic range.

[0047] It should be noted that after data cleaning and stability testing, key operation and maintenance data that meets the standards are obtained, which provides a data basis for subsequent operation and maintenance alarm analysis, thereby improving the accuracy of operation and maintenance alarms and reducing the false alarm rate of operation and maintenance alarms.

[0048] In an optional embodiment, based on the detection results, a first operation and maintenance data set is determined from the processed operation and maintenance data set, including: based on the detection results, determining target operation and maintenance data from the processed operation and maintenance data set, and eliminating the target operation and maintenance data to obtain eliminated operation and maintenance data, wherein the target operation and maintenance data is the operation and maintenance data corresponding to a time series that is not a stationary time series; and forming the first operation and maintenance data set based on the eliminated operation and maintenance data.

[0049] In some embodiments, in the process of determining the first operation and maintenance data set from the processed operation and maintenance data set based on the detection results, first, based on the detection results, the operation and maintenance data corresponding to the unstable time series (i.e., the target operation and maintenance data) can be determined from the processed operation and maintenance data set, and then the operation and maintenance data corresponding to the unstable time series are eliminated to obtain the eliminated operation and maintenance data, and the first operation and maintenance data set is formed based on the eliminated operation and maintenance data.

[0050] It should be noted that through the above process, key operation and maintenance data that meets the standards are obtained, which provides a data basis for subsequent operation and maintenance alarm analysis, thereby improving the accuracy of operation and maintenance alarms and reducing the false alarm rate of operation and maintenance alarms.

[0051] In an optional embodiment, feature extraction processing is performed on the first operation and maintenance data set through principal component analysis to obtain multiple operation and maintenance data features, including: performing data dimensionality reduction processing on the operation and maintenance data in the first operation and maintenance data set through principal component analysis to obtain a reduced-dimensional operation and maintenance data set; and performing feature extraction processing on the reduced-dimensional operation and maintenance data set to obtain multiple operation and maintenance data features.

[0052] In some embodiments, in the process of performing feature extraction processing on the first operation and maintenance data set through principal component analysis to obtain multiple operation and maintenance data features, the operation and maintenance data in the first operation and maintenance data set is first subjected to data dimensionality reduction processing through principal component analysis to obtain a reduced-dimensionality operation and maintenance data set, and then feature extraction processing is performed on the reduced-dimensionality operation and maintenance data set to obtain multiple operation and maintenance data features.

[0053] For example, after data cleaning and stationarity testing, a set of M×N×T-dimensional operation and maintenance data is obtained, where M is the number of network elements, N is the number of screened indicators, and T is the time period. Principal component analysis is used for dimensionality reduction. Principal component analysis dimensionality reduction is divided into time dimensionality reduction and indicator dimensionality reduction. For time dimensionality reduction, two periodic features are selected from six characteristic indicators (minimum value Min, maximum value Max, upper quantile Q1, mean Mean, lower quantile Q3, median Median), and finally the hourly data is simplified to descriptive indicators in days, that is, the new indicator dimension is M×2N; indicator dimensionality reduction is similar to time dimensionality reduction. After the time dimensionality reduction is completed, principal component analysis is used to construct three comprehensive indicators (i.e., operation and maintenance data features) from the 2N-dimensional features for final clustering, that is, the indicator dimension used for the clustering algorithm is finally simplified to M×3.

[0054] It should be noted that the above process reduces the amount of data calculation required for subsequent clustering and improves the efficiency of operation and maintenance alarm analysis.

[0055] In an optional embodiment, feature selection processing is performed on multiple operation and maintenance data features using a K-means clustering algorithm to obtain a target operation and maintenance data set, including: determining the initial number of clusters N based on the CH indicator, and randomly generating N initial cluster centers using the K-means clustering algorithm, where N is a positive integer; calculating the distance between each operation and maintenance data feature and the N initial cluster centers, and determining the cluster corresponding to each operation and maintenance data feature based on the distance; iterative clustering is performed based on the cluster clusters until a preset iteration termination condition is met to obtain the target operation and maintenance data set.

[0056] In some embodiments, in the process of performing feature selection processing on multiple operation and maintenance data features through the K-means clustering algorithm to obtain a target operation and maintenance data set, the initial number of clusters N is first determined based on the CH indicator, and N initial cluster centers are randomly generated through the K-means clustering algorithm. Then, the distance between each operation and maintenance data feature and the N initial cluster centers is calculated, and the cluster cluster corresponding to each operation and maintenance data feature is determined based on the distance. Then, iterative clustering is performed based on the cluster cluster until the preset iteration termination condition is met (for example, the left and right clusters no longer change) to obtain the target operation and maintenance data set.

[0057] In some embodiments, the CH (Calinski-Harabasz) metric is first used to determine the optimal number of clusters (i.e., the initial number of clusters N), and multiple clustering attempts are then performed to obtain the most stable clustering results. The CH metric describes compactness using an intra-class dispersion matrix and an inter-class dispersion matrix, respectively. A larger CH value indicates a more compact cluster and a more dispersed cluster, which translates to a better clustering result.

[0058] In some embodiments, after the optimal number of clusters is given, N initial cluster centers are randomly generated through the K-means clustering algorithm, and the data in the data set are divided into clusters according to their distance from the cluster centers. The average value of the data in each cluster is then taken as the new cluster center, and the above steps are repeated until the left and right clusters no longer change, thereby obtaining the target operation and maintenance data set.

[0059] Figure 2 is a flow chart of an optional data preprocessing process according to an embodiment of the present application. As shown in Figure 2, data collection is first performed to obtain an operation and maintenance data set, and then data cleaning and single-indicator time series data stationarity test are performed. Differential processing is performed on non-stationary time series data. For stationary time series data, a 24-hour period vector X is extracted, and Z is constructed based on the characteristics of the single-indicator period vector X. Then, a comprehensive vector V is obtained by principal component analysis of the full field Z. Then, the covariance matrix between indicators is examined, and new dimensions are named and low-dimensional visualization analysis of sample distribution is performed. Then, abnormal samples are deleted through the KS test distribution, and clustering is performed after the CH indicator is calculated to determine the optimal number of clusters with the K value to obtain clustering results. The various results are interpreted in combination with the meaning of the new dimension indicators.

[0060] It should be noted that in the embodiment of the present application, a method of data preprocessing based on quadratic interpolation and elimination of missing rate is adopted. First, the operation and maintenance data set to be processed is obtained, and the operation and maintenance data in the operation and maintenance data set to be processed is cleaned by the target algorithm to obtain a processed operation and maintenance data set. Then, the time series of the operation and maintenance data in the processed operation and maintenance data set is tested for stationarity to obtain a test result. Based on the test result, a first operation and maintenance data set is determined from the processed operation and maintenance data set. Then, feature extraction processing is performed on the first operation and maintenance data set by the principal component analysis method to obtain multiple operation and maintenance data features. Feature selection processing is performed on multiple operation and maintenance data features by the K-means clustering algorithm to obtain a target operation and maintenance data set. Then, an alarm analysis is performed based on the target operation and maintenance data set to obtain an alarm analysis result. Among them, the target algorithm includes at least a quadratic interpolation algorithm, the test result is used to characterize whether the time series is a stationary time series, and the alarm analysis result is used to indicate whether an alarm is issued.

[0061] In the above process, the operation and maintenance data in the operation and maintenance data set to be processed are cleaned by the quadratic interpolation algorithm, and the time series of the operation and maintenance data in the processed operation and maintenance data set are tested for stationarity, which can effectively improve the accuracy of data preprocessing, reduce the interference of abnormal data on subsequent alarm analysis, and reduce the errors caused by abnormal data. Then, the data set that has undergone data cleaning and stationarity testing is subjected to feature generation and selection, so that alarm analysis can be performed based on the target operation and maintenance data set, effectively improving the accuracy of operation and maintenance alarms and reducing the false alarm rate of operation and maintenance alarms.

[0062] It can be seen that the technical solution of the present application has achieved the goal of reducing the false alarm rate of operation and maintenance alarms, thereby achieving the technical effect of improving the accuracy of operation and maintenance alarms, and further solved the technical problem in the prior art of performing data preprocessing based on fixed rules, resulting in low accuracy of data preprocessing and resulting in low accuracy of operation and maintenance alarms.

[0063] Example 2

[0064] According to an embodiment of the present application, an embodiment of an operation and maintenance alarm device is provided, wherein FIG3 is a schematic diagram of an optional operation and maintenance alarm device according to an embodiment of the present application. As shown in FIG3 , the device includes: an acquisition module 301, configured to acquire an operation and maintenance data set to be processed, and perform data cleaning processing on the operation and maintenance data in the operation and maintenance data set to be processed by a target algorithm to obtain a processed operation and maintenance data set, wherein the target algorithm at least includes a quadratic interpolation algorithm; a determination module 302, configured to perform a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, and determine a first operation and maintenance data set from the processed operation and maintenance data set based on the test result, wherein the test result is used to characterize whether the time series is a stationary time series; a first processing module 303, configured to perform feature extraction processing on the first operation and maintenance data set by a principal component analysis method to obtain multiple operation and maintenance data features, and perform feature selection processing on the multiple operation and maintenance data features by a K-means clustering algorithm to obtain a target operation and maintenance data set; a second processing module 304, configured to perform alarm analysis based on the target operation and maintenance data set to obtain an alarm analysis result, wherein the alarm analysis result is used to indicate whether an alarm is issued.

[0065] It should be noted that the above-mentioned acquisition module 301, determination module 302, first processing module 303 and second processing module 304 correspond to steps S101 to S104 in the above-mentioned embodiment. The examples and application scenarios implemented by the four modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment 1.

[0066] In some embodiments, the acquisition module includes: a calculation module, used to calculate the missing rate of each operation and maintenance data in the operation and maintenance data set to be processed; a third processing module, used to eliminate the operation and maintenance data with a missing rate greater than the missing rate threshold from the operation and maintenance data set to be processed based on the missing rate of each operation and maintenance data, to obtain a second operation and maintenance data set; a fourth processing module, used to perform missing value filling processing on the operation and maintenance data in the second operation and maintenance data set through a quadratic interpolation algorithm to obtain a processed operation and maintenance data set.

[0067] In some embodiments, the fourth processing module includes: a first determination module, used to determine the missing values ​​of the operation and maintenance data in the second operation and maintenance data set, and determine the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment adjacent to the missing values; a fifth processing module, used to calculate the mean of the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment, and fill in the missing values ​​based on the mean to obtain the processed operation and maintenance data set.

[0068] In some embodiments, the determination module includes: a comparison module, which is used to calculate the probability value of the time series having a unit root through a unit root test algorithm, and compare the probability value with a preset threshold; a second determination module, which is used to detect that the time series is not a stationary time series as a detection result if the probability value is less than the preset threshold; and a third determination module, which is used to detect that the time series is a stationary time series as a detection result if the probability value is greater than or equal to the preset threshold.

[0069] In some embodiments, the determination module also includes: a fourth determination module, used to determine the target operation and maintenance data from the processed operation and maintenance data set based on the detection results, and eliminate the target operation and maintenance data to obtain the eliminated operation and maintenance data, wherein the target operation and maintenance data is the operation and maintenance data corresponding to the time series that is not a stationary time series; a generation module, used to form a first operation and maintenance data set based on the eliminated operation and maintenance data.

[0070] In some embodiments, the first processing module includes: a sixth processing module, used to perform data dimensionality reduction processing on the operation and maintenance data in the first operation and maintenance data set through principal component analysis to obtain a reduced-dimensional operation and maintenance data set; a seventh processing module, used to perform feature extraction processing on the reduced-dimensional operation and maintenance data set to obtain multiple operation and maintenance data features.

[0071] In some embodiments, the first processing module also includes: a fifth determination module, used to determine the initial number of clusters N based on the CH indicator, and randomly generate N initial cluster centers through the K-means clustering algorithm, where N is a positive integer; a sixth determination module, used to calculate the distance between each operation and maintenance data feature and the N initial cluster centers, and determine the cluster corresponding to each operation and maintenance data feature based on the distance; an eighth processing module, used to perform iterative clustering based on the cluster clusters until the preset iteration termination condition is met to obtain the target operation and maintenance data set.

[0072] Example 3

[0073] According to another aspect of an embodiment of the present application, a computer-readable storage medium is further provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned operation and maintenance alarm method when running.

[0074] Example 4

[0075] According to another aspect of an embodiment of the present application, an electronic device is also provided, which includes one or more processors; a memory for storing one or more programs, which enables the one or more processors to run the programs when the one or more programs are executed by the one or more processors, wherein the programs are configured to execute the above-mentioned operation and maintenance alarm method during operation.

[0076] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0077] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0078] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0079] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0080] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0081] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0082] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An operation and maintenance alarm method, comprising: Acquire an operation and maintenance data set to be processed, and perform data cleaning processing on the operation and maintenance data in the operation and maintenance data set to be processed by a target algorithm to obtain a processed operation and maintenance data set, wherein the target algorithm at least includes a quadratic interpolation algorithm; Performing a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, and determining a first operation and maintenance data set from the processed operation and maintenance data set according to the test result, wherein the test result is used to characterize whether the time series is a stationary time series; Performing feature extraction processing on the first operation and maintenance data set by using a principal component analysis method to obtain a plurality of operation and maintenance data features, and performing feature selection processing on the plurality of operation and maintenance data features by using a K-means clustering algorithm to obtain a target operation and maintenance data set; An alarm analysis is performed based on the target operation and maintenance data set to obtain an alarm analysis result, wherein the alarm analysis result is used to indicate whether to issue an alarm.

2. The method according to claim 1, wherein: The operation and maintenance data in the operation and maintenance data set to be processed is cleaned by a target algorithm to obtain a processed operation and maintenance data set, including: Calculating the missing rate of each operation and maintenance data in the operation and maintenance data set to be processed; According to the missing rate of each operation and maintenance data, operation and maintenance data with a missing rate greater than a missing rate threshold are removed from the operation and maintenance data set to be processed to obtain a second operation and maintenance data set; The operation and maintenance data in the second operation and maintenance data set is processed by filling missing values ​​using the quadratic interpolation algorithm to obtain the processed operation and maintenance data set.

3. The method according to claim 2, wherein: Performing missing value filling processing on the operation and maintenance data in the second operation and maintenance data set by using the quadratic interpolation algorithm to obtain the processed operation and maintenance data set includes: Determine a missing value of the operation and maintenance data in the second operation and maintenance data set, and determine the operation and maintenance data at a previous moment and the operation and maintenance data at a next moment adjacent to the missing value; The mean of the operation and maintenance data at the previous moment and the operation and maintenance data at the next moment is calculated, and the missing values ​​are filled according to the mean to obtain the processed operation and maintenance data set.

4. The method according to claim 1, wherein: Performing a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, including: Calculating the probability value of the time series having a unit root by a unit root test algorithm, and comparing the probability value with a preset threshold; If the probability value is less than the preset threshold, the time series is not a stationary time series as the detection result; If the probability value is greater than or equal to the preset threshold, the time series is regarded as a stationary time series as the detection result.

5. The method according to claim 1, characterized in that Determining a first operation and maintenance data set from the processed operation and maintenance data set according to the detection result includes: According to the detection result, target operation and maintenance data is determined from the processed operation and maintenance data set, and the target operation and maintenance data is eliminated to obtain the eliminated operation and maintenance data, wherein the target operation and maintenance data is the operation and maintenance data corresponding to the time series that is not a stationary time series; The first operation and maintenance data set is formed according to the eliminated operation and maintenance data.

6. The method according to claim 1, wherein: The first operation and maintenance data set is subjected to feature extraction processing by principal component analysis to obtain multiple operation and maintenance data features, including: Performing data dimensionality reduction processing on the operation and maintenance data in the first operation and maintenance data set by using the principal component analysis method to obtain a reduced-dimensional operation and maintenance data set; Perform feature extraction processing on the dimension-reduced operation and maintenance data set to obtain the multiple operation and maintenance data features.

7. The method according to claim 1, wherein: The multiple operation and maintenance data features are subjected to feature selection processing by using a K-means clustering algorithm to obtain a target operation and maintenance data set, including: Determine the initial number of clusters N according to the CH index, and randomly generate N initial cluster centers by the K-means clustering algorithm, where N is a positive integer; Calculating the distance between each operation and maintenance data feature and the N initial cluster centers, and determining the cluster corresponding to each operation and maintenance data feature according to the distance; Iterative clustering is performed according to the cluster clusters until a preset iteration termination condition is met to obtain the target operation and maintenance data set.

8. An operation and maintenance alarm device, comprising: An acquisition module is used to acquire an operation and maintenance data set to be processed, and perform data cleaning processing on the operation and maintenance data in the operation and maintenance data set to be processed by a target algorithm to obtain a processed operation and maintenance data set, wherein the target algorithm at least includes a quadratic interpolation algorithm; A determination module, configured to perform a stationarity test on the time series of the operation and maintenance data in the processed operation and maintenance data set to obtain a test result, and determine a first operation and maintenance data set from the processed operation and maintenance data set according to the test result, wherein the test result is used to characterize whether the time series is a stationary time series; A first processing module is used to perform feature extraction processing on the first operation and maintenance data set by a principal component analysis method to obtain a plurality of operation and maintenance data features, and perform feature selection processing on the plurality of operation and maintenance data features by a K-means clustering algorithm to obtain a target operation and maintenance data set; The second processing module is used to perform alarm analysis based on the target operation and maintenance data set to obtain an alarm analysis result, wherein the alarm analysis result is used to indicate whether to issue an alarm.

9. A computer-readable storage medium storing a computer program, wherein: The computer program is configured to execute the operation and maintenance alarm method described in any one of claims 1 to 7 when running.

10. An electronic device comprising one or more processors; a memory for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors are enabled to run the programs, wherein: The program is configured to execute the operation and maintenance alarm method described in any one of claims 1 to 7 when running.

Images