Fault point rapid positioning method and system based on electric power intelligent inspection system, and medium

Abstract

The invention discloses a fault point rapid positioning method and system based on an electric power intelligent inspection system and a medium, and the method comprises the steps: monitoring the electric energy quality data of each terminal transformer in real time, and calculating the real-time three-phase voltage unbalance degree based on the electric energy quality data of an alarm terminal transformer during a fault; then, searching is carried out in a fault abnormity library based on the real-time three-phase voltage unbalance degree, a fault topological relation is output, and m suspected fault terminal voltage devices are obtained based on a power grid topological graph; forming a time sequence by the power quality data of the alarm terminal voltage device before alarming and the m suspected fault terminal voltage devices, and inputting the time sequence into the fault positioning model to obtain a predicted fault terminal voltage device; in an electric power intelligent inspection system, on the basis of a WebGIS system, an Oracle Spatial technology is adopted to rapidly position and predict a fault terminal voltage device, and the fault terminal voltage device is displayed to maintenance personnel. According to the invention, the fault terminal voltage device is rapidly positioned and displayed based on the electric power intelligent inspection system, so that maintenance personnel can share fault point rapid positioning information in real time, the efficiency and accuracy of fault detection are effectively improved, and the practicability is good.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power intelligent inspection, and in particular relates to a method, system and medium for quickly locating a fault point based on an electric power intelligent inspection system. Background Art

[0002] Traditional manual inspections are conducted by workers holding detection equipment to collect equipment operation data. At present, the content of manual inspections is becoming more and more complicated, which has greatly reduced the quality of manual inspections and the timeliness of detection. The quality of manual inspections is highly random and is related to many factors such as the experience, sense of responsibility, and mental state of the staff, which often leads to misdetection, missed detection, and other situations, resulting in accidents. Manual inspections have the disadvantages of high labor intensity, low work efficiency, and high management costs.

[0003] With the rapid development of power system automation technology, the power intelligent inspection system has been widely promoted and applied. The power intelligent inspection system is a technical system that uses modern scientific and technological means such as computer technology, network technology, and software technology to monitor and inspect power grid lines and equipment in real time. The system improves the level of power grid operation and management through information and intelligent management. The power intelligent inspection system can meet the needs of on-site inspection personnel to query the design change of the inspection line, enter the completion data, and enable the inspection and maintenance personnel to query the work plan, equipment information and geographic information on site, record defects, and record the inspection status. It is convenient to grasp the changes in the operation status of the line and the surrounding environment, timely discover equipment defects and hidden dangers that endanger the safety of the line, and timely eliminate defects to prevent accidents. However, the topological structure of the distribution network is mostly radial, and the main line is mostly with branches and branch lines, which increases the difficulty of troubleshooting for power supply station employees. Once a power outage occurs, especially in bad weather, it takes 2-3 hours to find the fault point, and it takes 5-6 hours to find the fault point, which seriously affects the normal power supply of residents. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system and medium for quickly locating fault points based on an electric power intelligent inspection system, aiming to solve the above-mentioned problems.

[0005] The present invention is mainly achieved through the following technical solutions: The method for quickly locating a fault point based on an electric power intelligent inspection system comprises the following steps: Step S1: construct a power grid topology map, divide the power grid into regions according to the power grid topology map, and obtain power grid region i, where i=1, 2, ..., n, and n is a positive integer; Step S2: collecting historical data of each terminal voltage device in the power grid area i based on the power intelligent inspection system; Step S3: constructing a fault anomaly library based on the three-phase voltage imbalance range value and fault topology relationship when the alarm terminal voltage device fails in the historical data; Step S4: The power quality data of the alarm terminal voltage, the fault terminal voltage and the suspected fault terminal voltage at the time of the alarm and in the period before the alarm are combined into a time series, and the fault location and the fault state or normal state are marked, and the marked data samples are used to train the fault location model; Step S5: monitor the power quality data of each terminal transformer in real time, and calculate the real-time three-phase voltage imbalance based on the power quality data of the terminal transformer alarm when the fault occurs; then, search in the fault anomaly library based on the real-time three-phase voltage imbalance, output the fault topology relationship, and obtain m suspected fault terminal transformers based on the power grid topology diagram; Step S6: The power quality data of the alarm terminal voltage and m suspected fault terminal voltages in a period before the alarm are combined into a time series, and input into the fault prediction model to obtain the predicted fault terminal voltage; Step S7: In the power intelligent inspection system, based on the WebGIS system, Oracle Spatial technology is used to quickly locate and predict the faulty terminal voltage device, and display it to the maintenance personnel.

[0006] In order to better realize the present invention, further, in the step S2, the historical data includes the three-phase voltage imbalance when the alarm terminal voltage device fails, the fault type, the fault terminal voltage device, the fault topology relationship between the alarm terminal voltage device and the fault terminal voltage device, the suspected fault terminal voltage device with the same fault topology relationship, and the power quality data of the alarm terminal voltage device, the fault terminal voltage device and the suspected fault terminal voltage device in the cycle before the alarm.

[0007] In order to better implement the present invention, further, in the step S3, the three-phase voltage imbalance and the fault topology relationship when the alarm terminal voltage device fails are obtained, and the three-phase voltage imbalance range when the alarm terminal voltage device fails is obtained based on the fault topology relationship.

[0008] In order to better implement the present invention, further, in step S7, the power quality data of the predicted fault terminal voltage device is input into the fault prediction model to obtain the predicted fault type, and the corresponding maintenance plan is obtained based on the knowledge graph of fault type-maintenance plan.

[0009] In order to better implement the present invention, further, in the step S7, a fault prediction model is trained based on the power quality data of the fault terminal transformer and the fault type; and a knowledge graph of fault type-maintenance plan is constructed based on the fault type.

[0010] The present invention is mainly achieved through the following technical solutions: The fault point rapid positioning system based on the power intelligent inspection system is based on the above method and is characterized by including a historical database and a real-time data acquisition module, a fault anomaly analysis module, a fault positioning module and a map display module which are arranged in sequence.

[0011] A computer-readable storage medium stores a computer program, which implements the above method when executed by a processor.

[0012] The beneficial effects of the present invention are as follows: The present invention builds a fault anomaly library based on historical data and obtains a fault location model through training. In the fault anomaly library, the corresponding fault topological relationship is obtained based on the matching of the range value of the three-phase voltage imbalance degree, and the fault-involved area is deeply mined from the topological relationship of each terminal voltage device, that is, the suspected fault terminal voltage devices with the same topological relationship are obtained; then the power quality data of the alarm terminal voltage device before the alarm and the m suspected fault terminal voltage devices are formed into a time series, and the predicted fault terminal voltage device is obtained by rapid processing through the fault location model; finally, based on the WebGIS system and Oracle Spatial technology, the predicted fault terminal voltage device is rapidly located and displayed, so that each maintenance personnel can share the rapid location information of the fault point in real time, which effectively improves the efficiency and accuracy of fault detection and has good practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of the method for quickly locating a fault point based on an intelligent power inspection system in Example 1. DETAILED DESCRIPTION

[0014] Embodiment 1: The method for quickly locating fault points based on the power intelligent inspection system mainly includes the following steps: 1. Data collection and synchronization: Real-time collection of power quality data of each terminal transformer and automatic synchronization to the power intelligent inspection system.

[0015] (1) Construct a power grid topology map based on the WebGIS system and use Oracle Spatial technology to construct a power grid map including each terminal node; divide the power grid into regions according to the power grid topology map to obtain power grid region i, where i = 1, 2, …, n, and n is a positive integer; (2) Collect historical data of each terminal voltage regulator in the power grid area i; the historical data includes the three-phase voltage imbalance degree, fault type, fault terminal voltage regulator, fault topology relationship between the alarm terminal voltage regulator and the fault terminal voltage regulator, suspected fault terminal voltage regulators with the same fault topology relationship, and power quality data of the alarm terminal voltage regulator, fault terminal voltage regulator and suspected fault terminal voltage regulator in the cycle before the alarm. The power quality data includes current, voltage, temperature and phase.

[0016] (3) Establish a data table of the three-phase voltage imbalance and fault topology relationship when the alarm terminal voltage device fails, and obtain the three-phase voltage imbalance range when the alarm terminal voltage device fails based on the fault topology relationship, and construct a fault anomaly library containing the three-phase voltage imbalance range value and fault topology relationship when the alarm terminal voltage device fails based on the historical data; (4) The power quality data of the alarm terminal voltage, the fault terminal voltage, and the suspected fault terminal voltage at the time of the alarm and in the period before the alarm are combined into a time series, and the fault location and fault state or normal state are marked. The marked data samples are used to train the fault location model. The fault location model is trained through an end-to-end training strategy; the fault prediction model is trained based on the power quality data and fault type of the fault terminal voltage; and a knowledge graph of the fault type-maintenance plan is constructed based on the fault type; (5) Monitor the power quality data of each terminal transformer in real time, and calculate the real-time three-phase voltage imbalance based on the power quality data of the terminal transformer alarm when the fault occurs; then, search in the fault anomaly library based on the real-time three-phase voltage imbalance, output the fault topology relationship, and then obtain m suspected faulty terminal transformers based on the power grid topology diagram; (6) The power quality data of the pre-alarm terminal voltage and the m suspected fault terminal voltages are formed into a time series and input into the fault location model to obtain the predicted fault terminal voltage; the power quality data of the predicted fault terminal voltage is input into the fault prediction model to obtain the predicted fault type, and the corresponding maintenance plan is obtained based on the knowledge graph of fault type-maintenance plan; (7) In the power intelligent inspection system, based on the WebGIS system and using Oracle Spatial technology, the predicted fault terminal voltage is directly located and marked on the power grid map, and the fault type is marked with different colors and displayed to maintenance personnel.

[0017] 2. Fault identification and alarm: Use the power intelligent inspection system to read fault information and issue line fault alarm signals.

[0018] 3. Fault point location: Read the fault point location information, calculate the tower number near the fault point and the alarm terminal voltage device, and locate the line between the two.

[0019] 4. Plan generation and inspection: Based on the maintenance plan, the tools and equipment required for inspection are determined, and a plan is automatically generated for the operation and maintenance team. The operation and maintenance team formulates an inspection plan based on the plan, inspects the faulty line, and eliminates the fault.

[0020] Specifically, a time series is a series of data points indexed sequentially over time. A multivariate time series (MTS) represents multiple variables that depend on time, each of which is affected by past values ​​(called "time" dependency) and other variables (dimensions) based on their correlation. Based on the prediction-based anomaly detection method in time series, a prediction-based model can be trained to predict faulty terminal voltage transformers. Machine learning methods can be used for time series fault diagnosis: (1) Collect time series data; (2) Data preprocessing: cleaning and organizing data, dealing with missing values ​​and outliers; (3) Feature extraction: extracting valuable features from the original time series; (4) Model training and evaluation: Select a machine learning algorithm for model training and evaluation. For time series data, models such as random forest, XGBoost, or LSTM can be used.

[0021] Implement the prediction of faulty terminal voltage based on the trained model.

[0022] This embodiment applies WebGIS technology to the power intelligent inspection system. Through the developed network platform, the distribution network information and geographic information are published on the Web, which will greatly improve the efficiency of power system information management. The SCADA information is transmitted to the WebGIS system using Internet technology, so that power departments in different regions can share information data such as power equipment and operating conditions on a unified platform. Oracle Spatial technology includes data types for storing vector data types, raster data types, and persistent topological data. It manages spatial data through spatial data fields, metadata tables, and spatial indexes, and provides a series of query and spatial analysis packages on this basis for in-depth development. Each power spatial element in the system (such as cable lines, terminal equipment, etc.) is represented as a vector object with spatial characteristics and a set of attributes, and is stored through Oracle Spatial fields in Oracle DBA.

[0023] The present invention builds a fault anomaly library based on historical data and obtains a fault location model through training. In the fault anomaly library, the corresponding fault topological relationship is obtained based on the matching of the range value of the three-phase voltage imbalance degree, and the fault-involved area is deeply mined from the topological relationship of each terminal voltage device, that is, the suspected fault terminal voltage devices with the same topological relationship are obtained; then the power quality data of the alarm terminal voltage device before the alarm and the m suspected fault terminal voltage devices are formed into a time series, and the predicted fault terminal voltage device is obtained by rapid processing through the fault location model; finally, based on the WebGIS system and Oracle Spatial technology, the predicted fault terminal voltage device is rapidly located and displayed, so that each maintenance personnel can share the rapid location information of the fault point in real time, which effectively improves the efficiency and accuracy of fault detection and has good practicality.

[0024] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification or equivalent change made to the above embodiment based on the technical essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for quickly locating fault points based on an intelligent power inspection system, characterized in that: The following steps are involved: Step S1: construct a power grid topology map, divide the power grid into regions according to the power grid topology map, and obtain power grid region i, where i=1, 2, ..., n, and n is a positive integer; Step S2: collecting historical data of each terminal voltage device in the power grid area i based on the power intelligent inspection system; Step S3: constructing a fault anomaly library based on the three-phase voltage imbalance range value and fault topology relationship when the alarm terminal voltage device fails in the historical data; Step S4: The power quality data of the alarm terminal voltage, the fault terminal voltage and the suspected fault terminal voltage at the time of the alarm and in the period before the alarm are combined into a time series, and the fault location and the fault state or normal state are marked, and the marked data samples are used to train the fault location model; Step S5: monitor the power quality data of each terminal transformer in real time, and calculate the real-time three-phase voltage imbalance based on the power quality data of the terminal transformer alarm when the fault occurs; then, search in the fault anomaly library based on the real-time three-phase voltage imbalance, output the fault topology relationship, and obtain m suspected fault terminal transformers based on the power grid topology diagram; Step S6: The power quality data of the pre-alarm terminal voltage and the m suspected fault terminal voltages are combined into a time series, and input into a fault location model to obtain a predicted fault terminal voltage; Step S7: In the power intelligent inspection system, based on the WebGIS system, Oracle Spatial technology is used to quickly locate and predict the faulty terminal voltage device, and display it to the maintenance personnel.

2. The method for quickly locating a fault point based on an electric power intelligent inspection system according to claim 1 is characterized in that: In step S2, the historical data includes the three-phase voltage imbalance when the alarm terminal voltage device fails, the fault type, the fault terminal voltage device, the fault topology relationship between the alarm terminal voltage device and the fault terminal voltage device, the suspected fault terminal voltage device with the same fault topology relationship, and the power quality data of the alarm terminal voltage device, the fault terminal voltage device and the suspected fault terminal voltage device in the cycle before the alarm.

3. The method for quickly locating a fault point based on an electric power intelligent inspection system according to claim 2 is characterized in that: In the step S3, the three-phase voltage imbalance degree and the fault topology relationship when the alarm terminal voltage device fails are obtained, and the three-phase voltage imbalance degree range when the alarm terminal voltage device fails is obtained based on the fault topology relationship.

4. The method for quickly locating a fault point based on an electric power intelligent inspection system according to claim 2 is characterized in that: In the step S7, the power quality data of the predicted fault terminal voltage is input into the fault prediction model to obtain the predicted fault type, and the corresponding maintenance plan is obtained based on the knowledge graph of fault type-maintenance plan.

5. The method for quickly locating a fault point based on an electric power intelligent inspection system according to claim 4 is characterized in that: In the step S7, a fault prediction model is trained based on the power quality data and fault type of the fault terminal voltage device; Construct a knowledge graph of fault type-maintenance plan based on fault type.

6. A fault point rapid positioning system based on an electric power intelligent inspection system, based on the method described in any one of claims 1 to 5, characterized in that: It includes a historical database and a real-time data acquisition module, a fault anomaly analysis module, a fault location module and a map display module which are arranged in sequence.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method described in any one of claims 1 to 5 is implemented.

Images