A switchgear abnormality warning method and system

Abstract

The present invention discloses a switchgear abnormality early warning method and system, comprising: extracting the signal timing process of each control event according to the remote control information table associating the remote signal change table; processing the signal data to form a signal timing interval data set, calculating the timing of the signal experienced by each control event and the time interval between the signals; building a switchgear abnormality early warning model based on SVM, realizing early warning of potential abnormalities of the equipment by inputting the timing relationship of the control signal. The present invention reduces the probability of system accidents and improves the overall safety and reliability of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal state prediction of power system control signals, and in particular to a method and system for abnormal early warning of switch equipment. Background Art

[0002] In the power system, the control signal anomaly prediction of substation equipment is a key technology with important practical value. With the continuous development and improvement of the intelligence level of the power system, the substation, as an important part of the power system, the stable operation of its equipment is crucial to the safety and reliability of the entire power system. The abnormal control signal of substation equipment may be caused by a variety of factors, including equipment aging, external faults, operating errors, etc. If these anomalies are not discovered and handled in time, they may lead to equipment performance degradation, failures, and even system crashes. Therefore, it is particularly urgent to predict the abnormal control signals of substation equipment through advanced technical means. Summary of the invention

[0003] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a switch device abnormality warning method and system to solve the problem that the prior art cannot identify potential equipment problems, resulting in an increased risk of equipment failure.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for abnormal warning of a switch device, comprising:

[0008] According to the remote control information table, the remote signal change table is associated to extract the signal timing process of each control event;

[0009] Process the signal data to form a signal timing interval data set, and calculate the timing of the signal experienced by each control event and the time interval between the signals;

[0010] Build an abnormal warning model for switch equipment based on SVM, and realize early warning of potential abnormalities of equipment through the timing relationship of input control signals.

[0011] As a preferred solution of the switchgear abnormality early warning method of the present invention, wherein:

[0012] The process of extracting the signal timing of each control event includes the following steps:

[0013] Read the specified years data of the remote control table, format the occurrence time field of each record in a unified standard date, add a serial number id to the first column of each record, increase from 0 to 1, and add a timestamp corresponding to the occurrence time at the end of each record. Group the remote control signals according to whether USER_ID is 0, and store each group of signals with the plant id and interval id as the key value;

[0014] When traversing all key values ​​and the signals corresponding to the corresponding key values, extract the signal experienced by each control event according to the timestamp, sequence number and Userld value, and store it in Normal_Signal and Abnormal_Signal according to the signal status value;

[0015] Read the specified years data of the telemetering table, format the occurrence time, insert the sequence number in the first column and the timestamp in the last column, group the telemetering signals according to whether Userld is 0, and store each group of signals with the plant id and interval id as the key value;

[0016] Traverse the two sets of Normal_Signal and Abnormal_Signal, query and add the action status information of related equipment in the telesignal change table according to the ST_ID, BAY_ID and timestamp of each event control record, and form a complete control event signal timing process.

[0017] As a preferred solution of the switchgear abnormality early warning method of the present invention, wherein:

[0018] The step of grouping the remote control signals according to whether USER_ID is 0 includes:

[0019] When USER_ID is 0, it means no manual operation is required and is stored in the User_Id_Zero_Signal collection;

[0020] When USER_ID is not 0, it means that manual operation is required and stored in the User_Id_Non_Zero_Signal collection.

[0021] As a preferred solution of the switchgear abnormality early warning method of the present invention, wherein:

[0022] The storing of each group of signals using the plant area id and the interval id as key values ​​includes:

[0023] Traverse the User_Id_Zero_Signal collection, use the record's "ST_ID" plus "BAY_ID" as the key value, and store all records in the collection into the Map collection according to the corresponding key value;

[0024] Traverse the key values ​​of the Map collection and the operation records corresponding to the key values;

[0025] The recorded data whose first column sequence number difference is less than 2 and whose occurrence timestamp difference is less than 10 seconds is regarded as a control event;

[0026] Determine whether there is a record with STATUS of 6 or 10 in this event;

[0027] If STATUS is 6 or 10, it means control failed, and the event record is added to the Normal_Signal collection;

[0028] If STATUS is not 6 or 10, it means the control is normal, and the event record is added to the Abnormal_Signal collection;

[0029] As a preferred solution of the switchgear abnormality early warning method of the present invention, wherein:

[0030] The storing of each group of signals using the plant area id and the interval id as key values ​​also includes:

[0031] Traverse the User_Id_Non_Zero_Signal collection, use the record's "ST_ID" plus "BAY_ID" as the key value, and store all records in the collection into the _Map collection according to the corresponding key value;

[0032] Traverse the key values ​​of the _Map collection and the operation records corresponding to the key values;

[0033] The record data with the first column serial number difference less than 2 and the same USER_ID is regarded as a control event;

[0034] Determine whether there is a record with STATUS of 6 or 10 in this event;

[0035] If STATUS is 6 or 10, it means control failed, and the event record is added to the Normal_Signal collection;

[0036] If STATUS is not 6 or 10, it means the control is normal and the event record is added to the Abnormal_Signal collection.

[0037] As a preferred solution of the switchgear abnormality early warning method of the present invention, wherein:

[0038] The processing of the signal data to form a signal time series interval data set includes the following steps:

[0039] Traverse the Normal_Signal and Abnormal_Signal collections and delete useless related fields;

[0040] Calculate the signal time interval in each control event by calculating the time stamp difference between the signals;

[0041] The signal quantity, signal time interval array, signal status array and label of the control event are stored in Train_Data collection in array form; the data of the label in Normal_Signal is marked as 1, and the data of the label in Abnormal_Signal is marked as 0;

[0042] Process the training set Train_Data data, use the label field in the Train_Data data as the label data for model training, and the rest as the feature set features for model training;

[0043] Sum the array data of the signal time interval array and signal status array in Train_Data to obtain the training set processing result;

[0044] As a preferred solution of the switchgear abnormality early warning method of the present invention, wherein:

[0045] The construction of the switchgear abnormality warning model based on SVM includes:

[0046] The new control event signal data is processed into test samples according to the generated event data and the formed signal time series interval data set, and input into the SVM model for training to determine whether the equipment has abnormalities and issue early warnings for equipment with abnormal risks;

[0047] If the final model output is 0, it means that the device is abnormal;

[0048] If the final model output is 1, it means the device is normal.

[0049] In a second aspect, the present invention provides a switchgear abnormality warning system, comprising:

[0050] The extraction module associates the remote signal change table with the remote control information table to extract the signal timing process of each control event;

[0051] The processing module processes the signal data to form a signal timing interval data set, and calculates the timing of the signal experienced by each control event and the time interval between the signals;

[0052] Build a module to build an abnormal warning model for switch equipment based on SVM, and realize early warning of potential abnormalities of equipment through the timing relationship of input control signals.

[0053] In a third aspect, the present invention provides a computing device, comprising:

[0054] Memory, used to store programs;

[0055] The processor is used to execute the computer executable instructions, and the computer executable instructions, when executed by the processor, implement the steps of the switch device abnormality warning method.

[0056] In a fourth aspect, the present invention provides a computer-readable storage medium, comprising: when the program is executed by a processor, the steps of implementing the switching device abnormality warning method are implemented.

[0057] Beneficial effects of the present invention: The present invention successfully reduces the risk of equipment failure by detecting equipment anomalies in advance, effectively reducing the downtime and maintenance costs of the power system. Secondly, it optimizes the operation and maintenance efficiency, enables operation and maintenance personnel to carry out maintenance in a targeted manner, reduces routine inspection work, and improves the efficiency and accuracy of operation and maintenance work. In addition, the successful prediction of control signal anomalies reduces the probability of system accidents and improves the overall safety and reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0059] Figure 1 A basic flow chart of a switch device abnormality warning method provided by an embodiment of the present invention;

[0060] Figure 2 A schematic diagram of a data processing flow of a switch device abnormality warning method provided by an embodiment of the present invention;

[0061] Figure 3 A schematic diagram of a device abnormality warning process of a switch device abnormality warning method provided by an embodiment of the present invention;

[0062] Figure 4 A loss rate graph of an SVM model training for a switch device abnormality warning method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0064] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0065] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0066] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0067] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0068] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0069] Example 1

[0070] Reference Figure 1-3, is an embodiment of the present invention, and provides a switch device abnormality warning method, such as Figure 1 As shown, the following steps are included:

[0071] S1: According to the remote control information table, the remote signal change table is associated to extract the signal timing process of each control event;

[0072] Furthermore, Figure 2 As shown, it includes the following sub-steps:

[0073] S1.1 Query the operation records of the past 2 to 3 years from the remote control table. All records are sorted according to the timeline of occurrence, and the occurrence time field of each record is formatted with a unified standard date. Then, add a serial number id (increases from 0 to 1) to the first column of each record, and add a timestamp corresponding to the occurrence time at the end of each record;

[0074] S1.2 All records are divided into two categories according to whether the value of USER_ID (user ID) is 0. When USER_ID is 0, it indicates that no manual operation is required and the control operation is likely to be fine. On the contrary, when USER_ID is not 0, there is a high probability that there is a problem with the control operation. Traverse all records and store them in the User_Id_Zero_Signal and User_Id_Non_Zero_Signal collections respectively according to the above rules;

[0075] S1.3 traverses the User_Id_Zero_Signal collection, uses the record's "ST_ID (plant station ID)" plus "BAY_ID (interval ID)" as the key value, and stores all records in the collection into the Map collection according to the corresponding key value;

[0076] S1.4 Traverse the key values ​​of the Map collection, and then traverse the operation records corresponding to the key values. The recorded data with the first column sequence number difference less than 2 and the occurrence timestamp difference less than 10 seconds is regarded as a control event. Determine whether there is a record in this event whose STATUS (status value) is 6 or 10 (status 6 and 10 represent control failure). If such a record exists, add the event record to the Normal_Signal collection, otherwise add it to the Abnormal_Signal collection;

[0077] S1.5 traverses the User_Id_Non_Zero_Signal collection, uses the record's "ST_ID (plant station ID)" plus "BAY_ID (interval ID)" as the key value, and stores all records in the collection into the _Map collection according to the corresponding key value;

[0078] S1.6 Traverse the key values ​​of the _Map collection, and then traverse the operation records corresponding to the key values. The record data with the first column sequence number difference less than 2 and the same USER_ID is regarded as a control event. Determine whether there is a record in this event with a STATUS (status value) of 6 or 10 (status 6 and 10 represent control failure). If such a record exists, add the event record to the Normal_Signal collection, otherwise add it to the Abnormal_Signal collection;

[0079] S1.7 traverses the Normal_Signal and Abnormal_Signal collections, and queries and adds the action (or alarm) status information of the relevant equipment in the telesignal change table according to the ST_ID, BAY_ID and timestamp of each event control record, thereby forming a complete control event signal timing process. If the status of the telesignal change information queried in a certain event control record in Normal_Signal is 6 or 10, then the event control record is added with the relevant telesignal change information and transferred to Abnormal_Signal, and the data is deleted in Normal_Signal. After completing the above process, delete the useless related fields, including the first column number.

[0080] S2: Process the signal data to form a signal timing interval data set, and calculate the timing of the signal experienced by each control event and the time interval between the signals;

[0081] Furthermore, the following sub-steps are included:

[0082] S2.1 According to the S1 method, the signal timing process of each control event can be extracted, and these data are stored in the two sets of Normal_Signal and Abnormal_Signal. Traverse these two sets, first delete the useless related fields, and calculate the signal time interval in each control event (by calculating the timestamp difference between the signals), and store the number of signals, signal time interval array, signal status array, and label (the data in Normal_Signal is 1, and the data in Abnormal_Signal is 0) of this control event in the Train_Data collection in the form of an array;

[0083] S2.2 processes the training set Train_Data data, uses the label field in the Train_Data data as the label data for model training, and the rest as the feature set features for model training;

[0084] S2.3 sums the array data of the signal time interval array and the signal status array in Train_Data, normalizes them, and then randomly selects 20% of the data as the validation set Test_Data for model training.

[0085] S3: Build an abnormal warning model for switch equipment based on SVM, and realize early warning of potential abnormalities of equipment through the timing relationship of input control signals.

[0086] Furthermore, Figure 3 As shown, S3 contains the following:

[0087] Based on the signal timing interval data set formed by S2, we are committed to exploring the potential rules of the signal timing interval. By adopting the SVM (support vector machine) model, we build a switch equipment abnormality warning model. This model is based on the characteristic rules of the signal in the equipment control process, and aims to determine whether the equipment may be abnormal, so as to achieve the purpose of early warning.

[0088] Afterwards, the new control event signal data is processed into test samples according to the methods described in S1 and S2, and then input into the SVM model for training. Finally, the model will output 0 (representing abnormality) or 1 (representing normality). This process allows the new data to be monitored in real time through the model and provides timely warnings when the probability of control signal abnormality is high.

[0089] This embodiment also provides a switch device abnormality warning system, including:

[0090] The extraction module associates the remote signal change table with the remote control information table to extract the signal timing process of each control event;

[0091] The processing module processes the signal data to form a signal timing interval data set, and calculates the timing of the signal experienced by each control event and the time interval between the signals;

[0092] Build a module to build an abnormal warning model for switch equipment based on SVM, and realize early warning of potential abnormalities of equipment through the timing relationship of input control signals.

[0093] Furthermore, it also includes:

[0094] Memory, used to store programs;

[0095] A processor is used to load the program to execute the switch device abnormality early warning method.

[0096] This embodiment also provides a computer-readable storage medium storing a program, and when the program is executed by a processor, the switch device abnormality warning method is implemented.

[0097] The storage medium proposed in this embodiment and the switch device abnormality warning method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0098] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0099] Example 2

[0100] Reference Figure 4 Tables 1-6 are an embodiment of the present invention, which provides a method for abnormal warning of a switch device. In order to verify its beneficial effects, a scientific demonstration is carried out through specific implementation methods and implementation effects.

[0101] The present embodiment is as follows:

[0102] Control event generation: realize the process of obtaining the signal timing of each control event;

[0103] Signal timing interval data extraction and processing: Calculate the timing and time interval between each signal in the control event;

[0104] Establishing equipment abnormality warning model: Establishing support vector machine prediction model, and using the prediction model to judge whether there is a problem with the control signal.

[0105] It includes the following 3 steps:

[0106] Step 1: Automatically generate equipment control events based on remote control information and remote signal change signal data. The specific process includes the following sub-steps:

[0107] 1.1) Query the operation records of the past 2 to 3 years from the remote control table. All records are sorted according to the timeline of occurrence, and the occurrence time field of each record is formatted with a unified standard date. Then, add a serial number id (increases from 0 to 1) to the first column of each record, and add a timestamp corresponding to the occurrence time at the end of each record. The processed data is shown in Table 1;

[0108] Table 1. Results of processing the head and tail data of the remote control table

[0109] ID OCCUR_TIME … ST_ID BAY_ID … USER_ID … STATUS … TIMESTAMP 0 2022-01-04 20:56:25 113997366104686687 114278842239047301 0 18 1641300985.0 1 2022-01-04 20:56:26 113997366104686687 114278842239047301 0 18 1641300986.0 2 2022-01-04 20:56:26 113997366104686687 114278842239047301 0 22 1641300986.0 3 2022-01-04 21:58:52 113997366104686687 114278842239042818 0 2 1641304732.0 4 2022-01-04 21:58:54 113997366104686687 114278842239042818 0 2 1641304734.0 5 2022-01-04 21:58:54 113997366104686687 114278842239042818 0 5 1641304734.0 … … … … … … … … 30292 2024-01-04 11:07:05 113997366104686680 114278842239047241 0 18 1704337625.0 30293 2024-01-04 11:07:06 113997366104686680 114278842239047241 0 18 1704337626.0 30294 2024-01-04 11:07:06 113997366104686680 114278842239047241 0 22 1704337626.0

[0110] 1.2) Divide all records into two categories according to whether the value of USER_ID is 0. When USER_ID is 0, it means that no human operation is required and there is a high probability that there is no problem with the control operation. On the contrary, when USER_ID is not 0, there is a high probability that there is a problem with the control operation. Traverse all records and store them in the User_Id_Zero_Signal and User_Id_Non_Zero_Signal collections respectively according to the above rules;

[0111] 1.3) Traverse the User_Id_Zero_Signal collection, use the record's "ST_ID (plant station ID)" plus "BAY_ID (interval ID)" as the key value, and store all records in the collection into the Map collection according to the corresponding key value;

[0112] 1.4) Traverse the key values ​​of the Map collection, and then traverse the operation records corresponding to the key values. The record data with the first column sequence number difference less than 2 and the occurrence timestamp difference less than 10 seconds is regarded as a control event. Determine whether there is a record in this event with a STATUS (status value) of 6 or 10 (status 6 and 10 represent control failure). If such a record exists, add the event record to the Normal_Signal collection, otherwise add it to the Abnormal_Signal collection;

[0113] 1.5) Traverse the User_Id_Non_Zero_Signal collection, use the record's "ST_ID (plant station ID)" plus "BAY_ID (interval ID)" as the key value, and store all records in the collection into the _Map collection according to the corresponding key value;

[0114] 1.6) Traverse the key values ​​of the _Map collection, and then traverse the operation records corresponding to the key values. The record data with the first column sequence number difference less than 2 and the same USER_ID is regarded as a control event. Determine whether there is a record in this event whose STATUS (status value) is 6 or 10 (status 6 and 10 represent control failure). If such a record exists, add the event record to the Normal_Signal collection, otherwise add it to the Abnormal_Signal collection. Table 2 is the Normal_Signal collection data;

[0115] Table 2 Event signals stored in the Normal_Signal collection

[0116]

[0117] 1.7) Traverse the Normal_Signal and Abnormal_Signal collections, query and add the action (or alarm) status information of the relevant equipment in the telesignal change table (Table 3) according to the ST_ID, BAY_ID and timestamp of each event control record, so as to form a complete control event signal timing process. If the status of the telesignal change information queried in a certain event control record in Normal_Signal is 6 or 10, then the event control record is transferred to Abnormal_Signal after adding the relevant telesignal change information, and the data is deleted in Normal_Signal. Table 4 shows the data after adding the telesignal change information to Normal_Signal.

[0118] Table 3. Processing results of the head and tail data of the telemeter table

[0119]

[0120]

[0121] Table 4 Processing results of adding corresponding telesignal change signals to each event in Normal_Signal set

[0122]

[0123] Step 2: Extraction and processing of signal time interval data: Based on the event data generated by S1, extract the time interval data of all signals occurring during the evolution of a certain event, and process the data to form a signal time interval data set. The specific content includes the following sub-steps:

[0124] 2.1) According to the S1 method, the signal timing process of each control event can be extracted, including the first column number data, which are stored in the two sets of Normal_Signal and Abnormal_Signal. Traverse these two sets, first delete the useless related fields, and calculate the signal time interval in each control event (by calculating the timestamp difference between the signals), and store the number of signals, signal time interval array, signal status array, and label (the data in Normal_Signal is 1, and the data in Abnormal_Signal is 0) of this control event in the Train_Data collection in the form of an array;

[0125] 2.2) Process the training set Train_Data data, use the label field in the Train_Data data as the label data for model training, and the rest as the feature set features for model training;

[0126] 2.3) The array data of the signal time interval array and the signal status array in Train_Data are summed, and the results are shown in Table 5. Then, normalization is performed, and then 20% of the data is randomly intercepted as the verification set Test_Data for model training.

[0127] Table 5 Training set processing results

[0128] Label SIGNAL_NUM TOTAL_TIME_DIFF STATUS 0 1 4 7.0 7 1 1 4 6.0 7 2 1 4 14.0 91 3 1 3 1.0 9 4 1 8 10.0 121 … … … … … 11146 1 4 10.0 7 11147 1 3 1.0 6 11148 1 6 11.0 66 11149 1 5 9.0 64 11150 1 3 8.0 74

[0129] Step 3: Equipment abnormality warning method: Based on the signal timing interval data set formed by S2, the potential rules of the signal timing intervals are explored, and a switch equipment abnormality warning model is constructed. The recent signal data of the equipment is input to predict whether the equipment has abnormalities, and to issue warnings for equipment with abnormal risks.

[0130] The construction of the switchgear abnormality warning model according to S3 includes the following sub-steps:

[0131] Build SVM (support vector machine) early warning model:

[0132] SVM structure:

[0133] 1) Input Layer:

[0134] Number of nodes: 3 (depends on the number of input features).

[0135] 2) SVM model:

[0136] SVM is a linear model, so no hidden layer is needed here. The entire model is a linear classifier.

[0137] 3) Output Layer:

[0138] Number of nodes: 1 (two-classification problem)

[0139] Decision function: The decision function outputs the confidence or decision value that a sample belongs to the positive class.

[0140] The data processed in 2.3 is used as a training set to be input into the SVM model for training. After the training is completed, the obtained model is used for abnormal warning of switch equipment.

[0141] SVM model training and prediction are carried out according to the following steps:

[0142] 1) Data preparation: Prepare datasets for training and testing, and ensure that the datasets have been preprocessed and feature engineered.

[0143] 2) Model training: Use SVM to train the data and select appropriate kernel functions and other parameters for model training.

[0144] 3) Obtain output classification: SVM uses the predict method to obtain binary classification.

[0145] 4) Prediction decision: SVM output 0 represents abnormality and output 1 represents normality.

[0146] 5) Model evaluation: Use the test set to evaluate the integrated model and compare performance indicators such as accuracy, precision, recall, etc.

[0147] 6) Adjustment and optimization: Adjustment and optimization are performed based on the model evaluation results. The kernel function and parameters, regularization parameter C, and unbalanced data processing can be adjusted to further improve performance.

[0148] The model training results are as follows Figure 4 As shown, the training accuracy fully meets the prediction requirements.

[0149] The switchgear abnormality warning model constructed according to step 3 is characterized in that the new control event signal data is processed into a test sample according to the method described in step 1 and step 2, and then input into the SVM model for training, and the model will eventually output 0 (representing abnormality) or 1 (representing normality). This process allows the model to monitor new data in real time and provide timely warnings when the possibility of control signal abnormality is high.

[0150] It should be noted that Table 6 is a variable comparison table for the above scheme.

[0151] Table 6 Variable comparison table

[0152] USER_ID User ID field of the remote control table ST_ID The plant station id field of the remote control table BAY_ID Interval id field of the remote control table STATUS Status field of the remote control table User_Id_Zero_Signal Stores the set of signals with user ID 0 in the remote control table User_Id_Non_Zero_Signal Stores a collection of non-zero signals of user ID in the remote control table Normal_Signal Stores a collection of signals without abnormal events Abnormal_Signal A collection of abnormal event signals is stored

[0153] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A switchgear abnormality early warning method, characterized in that: include: According to the remote control information table, the remote signal change table is associated to extract the signal timing process of each control event; Process the signal data to form a signal timing interval data set, and calculate the timing of the signal experienced by each control event and the time interval between the signals; Build an abnormal warning model for switchgear based on SVM, and realize early warning of potential abnormalities of equipment by inputting the timing relationship of control signals; The process of extracting the signal timing of each control event includes the following steps: Read the specified years of data in the remote control information table, format the occurrence time field of each record in a unified standard date format, add a serial number id to the first column of each record, increase from 0 to 1, and add a timestamp corresponding to the occurrence time at the end of each record. Group the remote control signals according to whether USER_ID is 0, and store each group of signals with the plant id and interval id as the key value; When traversing all key values ​​and the signals corresponding to the corresponding key values, extract the signal experienced by each control event according to the timestamp, sequence number and Userld value, and store it in Normal_Signal and Abnormal_Signal according to the signal status value; Read the specified years data of the remote signal change table, format the occurrence time, insert the sequence number in the first column and the timestamp in the last column, group the remote control signals according to whether Userld is 0, and store each group of signals with the plant id and interval id as the key value; Traverse the two sets of Normal_Signal and Abnormal_Signal, query and add the action status information of related equipment in the telesignal change table according to the ST_ID, BAY_ID and timestamp of each event control record, and form a complete control event signal timing process.

2. The switchgear abnormality early warning method according to claim 1, characterized in that: The step of grouping the remote control signals according to whether USER_ID is 0 includes: When USER_ID is 0, it means no manual operation is required and is stored in the User_Id_Zero_Signal collection; When USER_ID is not 0, it means that manual operation is required and stored in the User_Id_Non_Zero_Signal collection.

3. The switchgear abnormality early warning method according to claim 2, characterized in that: The storing of each group of signals using the plant area id and the interval id as key values ​​includes: Traverse the User_Id_Zero_Signal collection, use the record's "ST_ID" plus "BAY_ID" as the key value, and store all records in the collection into the Map collection according to the corresponding key value; Traverse the key values ​​of the Map collection and the operation records corresponding to the key values; The recorded data whose first column sequence number difference is less than 2 and whose occurrence timestamp difference is less than 10 seconds is regarded as a control event; Determine whether there is a record with STATUS of 6 or 10 in this event; If STATUS is 6 or 10, it means control failed, and the event record is added to the Normal_Signal collection; If STATUS is not 6 or 10, it means the control is normal and the event record is added to the Abnormal_Signal collection.

4. The switchgear abnormality early warning method according to claim 3, characterized in that: The storing of each group of signals using the plant area id and the interval id as key values ​​also includes: Traverse the User_Id_Non_Zero_Signal collection, use the record's "ST_ID" plus "BAY_ID" as the key value, and store all records in the collection into the _Map collection according to the corresponding key value; Traverse the key values ​​of the _Map collection and the operation records corresponding to the key values; The record data with the first column serial number difference less than 2 and the same USER_ID is regarded as a control event; Determine whether there is a record with STATUS of 6 or 10 in this event; If STATUS is 6 or 10, it means control failed, and the event record is added to the Normal_Signal collection; If STATUS is not 6 or 10, it means the control is normal and the event record is added to the Abnormal_Signal collection.

5. The switchgear abnormality early warning method according to claim 4, characterized in that: The processing of the signal data to form a signal time series interval data set includes the following steps: Traverse the Normal_Signal and Abnormal_Signal collections and delete useless related fields; Calculate the time interval between signals in each control event by calculating the difference in timestamps between the signals; The signal quantity, signal time interval array, signal status array and label of the control event are stored in Train_Data collection in array form; the data of the label in Normal_Signal is marked as 1, and the data of the label in Abnormal_Signal is marked as 0; Process the training set Train_Data data, use the label field in the Train_Data data as the label data for model training, and the rest as the feature set features for model training; The array data of the signal time interval array and the signal status array in Train_Data are summed to obtain the training set processing result.

6. The switchgear abnormality early warning method according to claim 5, characterized in that: The construction of the switchgear abnormality warning model based on SVM includes: The new control event signal data is processed into test samples according to the generated event data and the formed signal time series interval data set, and input into the SVM model for training to determine whether the equipment has abnormalities and issue early warnings for equipment with abnormal risks; If the final model output is 0, it means that the device is abnormal; If the final model output is 1, it means the device is normal.

7. A switchgear abnormality warning system according to claim 1, characterized in that: The extraction module associates the remote signal change table with the remote control information table to extract the signal timing process of each control event; The processing module processes the signal data to form a signal timing interval data set, and calculates the timing of the signal experienced by each control event and the time interval between the signals; Build a module to build an abnormal warning model for switch equipment based on SVM, and realize early warning of potential abnormalities of equipment through the timing relationship of input control signals.

8. An electronic device, characterized in that: include: Memory, used to store programs; A processor is used to load the program to execute the steps of the switch device abnormality early warning method as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing a program, characterized in that: When the program is executed by a processor, the steps of the switch device abnormality early warning method according to any one of claims 1 to 7 are implemented.

Images