Optimal tripping point intelligent judgment method based on equipment topological relation

Abstract

The invention relates to an intelligent judgment method for an optimal tripping point based on an equipment topological relation, and belongs to the technical field of power grid dispatching. The method comprises the following five steps: establishing an equipment topological relation model, acquiring power grid data in real time, constructing the model, intelligently judging an optimal tripping point, and feeding back and executing. According to the method, the topological connection relation of the equipment in the power grid is comprehensively considered, the power grid data collected in real time is combined, and the tripping point is quickly and accurately judged by using an advanced intelligent algorithm, so that the operation safety and stability of the power grid are improved, the influence of faults on the power grid is reduced, and the reliability and continuity of power supply are guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of power grid dispatching, and in particular relates to an optimal trip point intelligent determination method based on equipment topology relationship. Background Art

[0002] During the operation and dispatching of power grids, tripping events occur from time to time. The current determination of tripping points mainly relies on manual judgment or simple threshold setting. These methods have problems such as inaccurate judgment and slow response. Especially in complex power grid structures, it is difficult to quickly and accurately find the best tripping point, which may lead to the expansion of fault range and the extension of recovery time.

[0003] The topological relationship of devices in the power grid refers to the connection relationship between various devices in the power grid (such as transformers, switches, lines, etc.). Most of the existing trip point determination methods are based on simple threshold judgments. For example, the current exceeding the set threshold is determined as a trip point. This method ignores the topological relationship between power grid devices, resulting in inaccurate determination results. In addition, although some methods take into account the connection relationship between devices, the determination process is complicated and difficult to promote in practical applications.

[0004] Therefore, how to overcome the shortcomings of existing technologies is an urgent problem to be solved in the field of power grid dispatching technology. Summary of the invention

[0005] The present invention aims to provide an optimal trip point intelligent determination method based on the topological relationship of the equipment to solve the problems of inaccurate determination, slow response speed and high complexity in the current power grid trip point determination process. By comprehensively considering the topological connection relationship of the equipment in the power grid, combining the real-time collected power grid data, and using advanced intelligent algorithms to make fast and accurate trip point determinations, the operational safety and stability of the power grid can be improved, the impact of faults on the power grid can be reduced, and the reliability and continuity of power supply can be guaranteed.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0007] An optimal trip point intelligent determination method based on equipment topology relationship comprises the following steps:

[0008] Step (1), establish a device topology relationship model:

[0009] The devices in the power grid are abstracted as nodes, and the connection relationship between devices is abstracted as edges, so as to establish the device topology relationship model of the power grid;

[0010] Step (2), real-time collection of power grid data:

[0011] Real-time collection of operating data and historical data of equipment in the power grid;

[0012] Step (3), model construction:

[0013] Establishing an optimal trip point calculation model based on the power grid topology model obtained in step (1) and the operation data obtained in step (2);

[0014] Step (4), intelligently determine the best trip point:

[0015] When a fault occurs, the optimal trip point calculation model is used to calculate the optimal trip point set, and the corresponding trip point location and the power supply range after the trip point trips are displayed in the topological relationship model;

[0016] Step (5), feedback and execution: the business personnel feed back the optimal trip point set to the power grid control system, execute the corresponding tripping operation, and record the tripping event at the same time.

[0017] Further, preferably, step (1) specifically comprises the following steps:

[0018] (1.1) Single-line diagram identification:

[0019] Get the single-line diagram of the power grid;

[0020] (1.2) Image preprocessing:

[0021] Preprocessing the single-line diagram of the acquired power grid;

[0022] (1.3) Image segmentation:

[0023] Divide the single-line diagram into multiple areas, each area corresponds to a device or a group of devices;

[0024] (1.4) Device abstraction:

[0025] Abstract the devices in the power grid as nodes in the graph;

[0026] (1.5) Connection relationship identification:

[0027] Identify the connection relationship between devices;

[0028] (1.6) Connection relationship abstraction:

[0029] Abstract the connection relationship between devices as edges in the graph;

[0030] (1.7) Graph structure construction:

[0031] Use graph structures to represent devices and their connections.

[0032] Further, preferably, in step (1.2), the preprocessing includes binarization and denoising.

[0033] Furthermore, preferably, in step (1.4), the equipment includes transformers, switches, circuit breakers, feeders, busbars, lines, capacitors and reactors.

[0034] Further, preferably, in step (1.5), the connection relationship is identified by way one or way two, specifically as follows:

[0035] Method 1: Proximity analysis: Analyze the relative positions between devices to determine whether they are connected;

[0036] Method 2: Path tracking: Starting from one device, trace along a line to another device to determine the connection relationship between them.

[0037] Furthermore, preferably, in step (2), the operating data includes the number of transformers, the scope of power outage, important protection equipment, asynchronous reclosing equipment, faulty equipment, tripping equipment, the number of users, and the number of topologies.

[0038] Further, preferably, the specific method of step (3) is:

[0039] (3.1) Data preprocessing:

[0040] Preprocess the operating data when historical faults occur to remove outliers and missing values, and perform normalization;

[0041] The operation data when the historical fault occurs includes the number of transformers affected, the area of ​​power outage, the number of important protection equipment, the number of asynchronous reclosing equipment, the number of faulty equipment, the number of tripped equipment, the number of users and the number of topologies when the historical fault occurs and after the trip;

[0042] (3.2) Data classification:

[0043] Classify the collected data into equipment faults and line faults according to the fault type;

[0044] (3.3) Optimal trip point calculation model model construction:

[0045] (3.3.1) Correlation analysis

[0046] Taking the power supply area as y, analyze the correlation between the eight influencing factors, including the number of transformers x1, the power outage area x2, the number of important protection equipment x3, the number of asynchronous reclosing equipment x4, the number of faulty equipment x5, the number of tripping equipment x6, the number of users x7, and the number of topologies x8, and the power supply area as y;

[0047] Specifically, for the two fault types, the Pearson correlation coefficient method is used to analyze the correlation between x1~x8 and y, and the Lasso regression is used to remove the collinearity problem between the influencing factors, so as to extract the key features that affect the power supply range for the two fault types;

[0048] (3.3.2) Construction of optimal trip point calculation model

[0049] The power grid is considered as a graphic plane, with the fault point as the starting point, the end of the line as the end point, and the nodes in the middle as the various tripping points;

[0050] For each trip point, calculate the parameter value of the trip point under the fault type;

[0051] The parameter value of the trip point is calculated as follows:

[0052] Multiply each key feature selected in (3.3.1) corresponding to the fault type by the weight of the key feature, sum the results, and obtain the parameter value of the trip point;

[0053] Under each fault type, the sum of the weights of all key features is 1;

[0054] The A* algorithm is used to find the minimum path of the sum of the parameter values ​​of the tripping points from the fault point to the end point, and the set of these tripping points is the optimal set of tripping points;

[0055] The optimal trip point set corresponding to the operating data when historical faults occurred is used to solve the weights of the key features under each fault type. The solution is used to bring the return trip point parameter value into the calculation, thereby obtaining the optimal trip point calculation model.

[0056] Furthermore, preferably, in (3.2), the fault types include line faults and equipment faults.

[0057] Furthermore, preferably, in (3.3.1), when using the Pearson correlation coefficient method to analyze the correlation between x1-x8 and y, the influencing factors with a correlation coefficient greater than 0.5 are selected, and then Lasso regression is performed.

[0058] The present invention provides an optimal trip point intelligent determination method based on the equipment topology relationship, which determines the optimal trip point when a fault occurs in the power grid through precise analysis and judgment, so as to reduce the fault range and ensure the stable operation of the power grid. Compared with the prior art, the technical effects of the present invention are mainly reflected in the following aspects:

[0059] (1) Improving judgment accuracy: By considering the topological relationship between power grid devices, the present invention can more accurately judge the fault point and its impact range, thereby determining the optimal tripping point and reducing the possibility of misjudgment and missed judgment.

[0060] (2) Shorten response time: With the help of real-time data acquisition and intelligent algorithm analysis, the present invention can quickly respond to power grid faults, achieve rapid fault location and intelligent determination of the optimal tripping point, and reduce the impact of faults on power grid operation.

[0061] (3) Enhanced grid stability: By accurately determining the optimal tripping point, the present invention helps to isolate the fault area in a timely manner when a fault occurs, preventing the fault range from expanding, thereby enhancing the stability and reliability of the grid.

[0062] (4) Improving operation and maintenance efficiency: The method provided by the present invention not only simplifies the process of trip point determination, but also reduces the need for manual intervention, thereby improving the efficiency and automation level of power grid operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 The present invention is a flow chart of the method for intelligently determining the optimal tripping point based on the equipment topology relationship. DETAILED DESCRIPTION

[0064] The present invention is further described in detail below in conjunction with embodiments.

[0065] Those skilled in the art will appreciate that the following examples are only used to illustrate the present invention and should not be considered to limit the scope of the present invention. If no specific techniques or conditions are specified in the examples, the techniques or conditions described in the literature in the art or the product specifications are used. If the manufacturer of the materials or equipment used is not specified, they are all conventional products that can be purchased.

[0066] Example 1

[0067] An optimal trip point intelligent determination method based on equipment topology relationship comprises the following steps:

[0068] Step (1), establish a device topology relationship model:

[0069] The devices in the power grid are abstracted as nodes, and the connection relationship between devices is abstracted as edges, so as to establish the device topology relationship model of the power grid;

[0070] Step (2), real-time collection of power grid data:

[0071] Real-time collection of operating data and historical data of equipment in the power grid;

[0072] Step (3), model construction:

[0073] Establishing an optimal trip point calculation model based on the power grid topology model obtained in step (1) and the operation data obtained in step (2);

[0074] Step (4), intelligently determine the best trip point:

[0075] When a fault occurs, the optimal trip point calculation model is used to calculate the optimal trip point set, and the corresponding trip point location and the power supply range after the trip point trips are displayed in the topological relationship model;

[0076] Step (5), feedback and execution: the business personnel feed back the optimal trip point set to the power grid control system, execute the corresponding tripping operation, and record the tripping event at the same time.

[0077] Example 2

[0078] An optimal trip point intelligent determination method based on equipment topology relationship comprises the following steps:

[0079] Step (1), establish a device topology relationship model:

[0080] The devices in the power grid are abstracted as nodes, and the connection relationship between devices is abstracted as edges, so as to establish the device topology relationship model of the power grid;

[0081] Step (2), real-time collection of power grid data:

[0082] Real-time collection of operating data and historical data of equipment in the power grid;

[0083] Step (3), model construction:

[0084] Establishing an optimal trip point calculation model based on the power grid topology model obtained in step (1) and the operation data obtained in step (2);

[0085] Step (4), intelligently determine the best trip point:

[0086] When a fault occurs, the optimal trip point calculation model is used to calculate the optimal trip point set, and the corresponding trip point location and the power supply range after the trip point trips are displayed in the topological relationship model;

[0087] Step (5), feedback and execution: the business personnel feed back the optimal trip point set to the power grid control system, execute the corresponding tripping operation, and record the tripping event at the same time.

[0088] The optimal trip point intelligent determination method based on device topology relationship, step (1) specifically includes the following steps:

[0089] (1.1) Single-line diagram identification:

[0090] Get the single-line diagram of the power grid;

[0091] (1.2) Image preprocessing:

[0092] Preprocessing the single-line diagram of the acquired power grid;

[0093] (1.3) Image segmentation:

[0094] Divide the single-line diagram into multiple areas, each area corresponds to a device or a group of devices;

[0095] (1.4) Device abstraction:

[0096] Abstract the devices in the power grid as nodes in the graph;

[0097] (1.5) Connection relationship identification:

[0098] Identify the connection relationship between devices;

[0099] (1.6) Connection relationship abstraction:

[0100] Abstract the connection relationship between devices as edges in the graph;

[0101] (1.7) Graph structure construction:

[0102] Use graph structures to represent devices and their connections.

[0103] In step (1.2), preprocessing includes binarization and denoising.

[0104] In step (1.4), the equipment includes transformers, switches, circuit breakers, feeders, busbars, lines, capacitors and reactors.

[0105] In step (1.5), the connection relationship is identified by method 1 or method 2, as follows:

[0106] Method 1: Proximity analysis: Analyze the relative positions between devices to determine whether they are connected;

[0107] Method 2: Path tracking: Starting from one device, trace along a line to another device to determine the connection relationship between them.

[0108] In step (2), the operation data includes the number of transformers, the scope of power outage, important protection equipment, asynchronous reclosing equipment, faulty equipment, tripping equipment, the number of users, and the number of topologies.

[0109] The specific method of step (3) is:

[0110] (3.1) Data preprocessing:

[0111] Preprocess the operating data when historical faults occur to remove outliers and missing values, and perform normalization;

[0112] The operation data when the historical fault occurs includes the number of transformers affected, the area of ​​power outage, the number of important protection equipment, the number of asynchronous reclosing equipment, the number of faulty equipment, the number of tripped equipment, the number of users and the number of topologies when the historical fault occurs and after the trip;

[0113] (3.2) Data classification:

[0114] Classify the collected data into equipment faults and line faults according to the fault type;

[0115] (3.3) Optimal trip point calculation model model construction:

[0116] (3.3.1) Correlation analysis

[0117] Taking the power supply area as y, analyze the correlation between the eight influencing factors, including the number of transformers x1, the power outage area x2, the number of important protection equipment x3, the number of asynchronous reclosing equipment x4, the number of faulty equipment x5, the number of tripping equipment x6, the number of users x7, and the number of topologies x8, and the power supply area as y;

[0118] Specifically, for the two fault types, the Pearson correlation coefficient method is used to analyze the correlation between x1~x8 and y, and the Lasso regression is used to remove the collinearity problem between the influencing factors, so as to extract the key features that affect the power supply range for the two fault types;

[0119] (3.3.2) Construction of optimal trip point calculation model

[0120] The power grid is considered as a graphic plane, with the fault point as the starting point, the end of the line as the end point, and the nodes in the middle as the various tripping points;

[0121] For each trip point, calculate the parameter value of the trip point under the fault type;

[0122] The parameter value of the trip point is calculated as follows:

[0123] Multiply each key feature selected in (3.3.1) corresponding to the fault type by the weight of the key feature, sum the results, and obtain the parameter value of the trip point;

[0124] Under each fault type, the sum of the weights of all key features is 1;

[0125] The A* algorithm is used to find the minimum path of the sum of the parameter values ​​of the tripping points from the fault point to the end point, and the set of these tripping points is the optimal set of tripping points;

[0126] The optimal trip point set corresponding to the operating data when historical faults occurred is used to solve the weights of the key features under each fault type. The solution is used to bring the return trip point parameter value into the calculation, thereby obtaining the optimal trip point calculation model.

[0127] In (3.2), the fault types include line faults and equipment faults.

[0128] In (3.3.1), when using the Pearson correlation coefficient method to analyze the correlation between x1~x8 and y, select the influencing factors with a correlation coefficient greater than 0.5, and then perform Lasso regression.

[0129] Example 3

[0130] An optimal trip point intelligent determination method based on equipment topology relationship comprises the following steps:

[0131] Step (1), establish a device topology relationship model:

[0132] The devices in the power grid (such as transformers, etc.) are abstracted as nodes, the connection relationships between devices are abstracted as edges, and a device topology relationship model of the power grid is established.

[0133] (1.1) Single-line diagram identification:

[0134] To prepare the single-line diagram data, you first need to obtain the single-line diagram data of the power grid. This data is usually contained in PDF documents, CAD drawings, SVG files or other formats. If the data exists in electronic form, it can be used directly; if it is a paper document, it needs to be digitized first.

[0135] (1.2) Image preprocessing:

[0136] If the single-line diagram exists in the form of an image, it needs to be preprocessed, including: Image sharpening: improving the resolution and contrast of the image to make the lines clearer. Noise removal: removing irrelevant noise in the image, such as background noise, text, etc.

[0137] (1.3) Image segmentation:

[0138] Divide the single line diagram into multiple areas, each area corresponds to a device or a group of devices. This step can be achieved through image processing techniques, such as edge detection, contour extraction, etc.

[0139] (1.4) Device abstraction:

[0140] First, the devices in the power grid are abstracted as nodes (vertices) in the graph. Common devices include but are not limited to: transformer, switch, circuit breaker, feeder, busbar, line, capacitor, and reactor.

[0141] (1.5) Connection relationship identification:

[0142] Identify the connection relationship between devices and abstract it as edges in the graph. The identification of connection relationship can be achieved through the following methods:

[0143] Proximity analysis: Analyzes the relative locations of devices to determine whether they are connected.

[0144] Path tracking: Starting from one device, follow the line to another device to determine the connection relationship between them, as follows:

[0145] (1.5.1) Edge detection:

[0146] Use edge detection algorithms (such as Canny edge detection, Sobel operator, etc.) to detect edges in the image. The results of edge detection can be used to extract the boundaries of devices and the lines connecting devices.

[0147] (1.5.2) Connected component analysis:

[0148] Connected component analysis is the process of isolating connected pixel regions in an image (i.e., regions consisting of pixels connected by the same color or gray level). This step can help us find the location of devices and their connecting lines.

[0149] (1.5.3) Starting point selection:

[0150] Select a device as the starting point, usually the center point or connection point of the device. The center point of the device can be obtained by calculating the geometric center of the device area.

[0151] (1.5.4) Path tracking:

[0152] Starting from the selected starting point, trace along the line to another device. The specific steps of path tracking are as follows:

[0153] (1) Find lines near the starting point: Start searching for lines near the starting point. You can use an 8-connected search to find lines around the starting point.

[0154] (2) Tracking lines: Once a line near the starting point is found, start tracking along the line. You can use one of the following two methods during the tracking process:

[0155] Direction tracking: Record the direction of the line and continue moving forward based on the direction information.

[0156] Boundary Tracking: Move along the boundary of the line until the boundary of the next device is found.

[0157] (3) Line segmentation: During the tracking process, when encountering a bifurcation point, it is necessary to determine the branch of the line. At this time, the bifurcation point detection algorithm can be used to identify the branch of the line and track each branch separately.

[0158] (4) Termination conditions: The termination conditions of path tracking include:

[0159] Reaching another device: Stop tracking when it reaches the boundary of another device.

[0160] Reaching the preset length: If the tracking length exceeds the preset threshold, it means that incorrect tracking may have occurred and the tracking can be stopped.

[0161] (1.6) Connection relationship abstraction:

[0162] The physical connections between devices are abstracted as edges in the graph. These connections reflect the electrical connections between devices, for example: a transformer is connected to a bus, a switch is connected to a line, and a line is connected to another line or bus.

[0163] (1.7) Graph structure construction:

[0164] Use a graph structure to represent devices and their connections. You can use a graph database (such as Neo4j) to build a topology structure.

[0165] Step (2), real-time collection of power grid data:

[0166] Scripts are written using Python technology to automatically collect operating data of power grid equipment in the power grid OCS system and operating system, including the number of transformers, power outage scope, important protection equipment, asynchronous reclosing equipment, faulty equipment, tripped equipment, and the number of users.

[0167] These data can be transmitted to the data analysis center via the network for subsequent processing.

[0168] Step (3), model construction:

[0169] According to the power grid topology model obtained in step (1) and the operation data obtained in step (2), an optimal trip point calculation model is established to analyze the optimal trip point device when a circuit breaker fault may occur in the power grid.

[0170] (3.1) Data preprocessing:

[0171] The operating data at the time of historical faults are preprocessed to remove abnormal values ​​and missing values, and normalized; the normalization adopts the maximum and minimum normalization method;

[0172] The operation data when the historical fault occurs includes the number of transformers affected, the area of ​​power outage, the number of important protection equipment, the number of asynchronous reclosing equipment, the number of faulty equipment, the number of tripped equipment, the number of users and the number of topologies when the historical fault occurs and after the trip;

[0173] (3.2) Data classification:

[0174] Classify the collected data into equipment faults and line faults according to the fault type;

[0175] (3.3) Optimal trip point calculation model construction

[0176] (3.3.1) Correlation analysis

[0177] Taking the power supply area as y, analyze the correlation between the eight influencing factors, including the number of transformers x1, the power outage area x2, the number of important protection equipment x3, the number of asynchronous reclosing equipment x4, the number of faulty equipment x5, the number of tripping equipment x6, the number of users x7, and the number of topologies x8, and the power supply area as y;

[0178] Specifically, for the two fault types, the Pearson correlation coefficient method is used to analyze the correlation between x1~x8 and y, and the Lasso regression is used to remove the collinearity problem between the influencing factors, so as to extract the key features that affect the power supply range for the two fault types;

[0179] (3.3.2) Construction of optimal trip point calculation model

[0180] The power grid is considered as a graphic plane, with the fault point as the starting point, the end of the line as the end point, and the nodes in the middle as the various tripping points;

[0181] For each trip point, calculate the parameter value of the trip point under the fault type;

[0182] The parameter value of the trip point is calculated as follows:

[0183] Multiply each key feature selected in (3.3.1) corresponding to the fault type by the weight of the key feature, sum the results, and obtain the parameter value of the trip point;

[0184] Under each fault type, the sum of the weights of all key features is 1;

[0185] The A* algorithm is used to find the minimum path of the sum of the parameter values ​​of the tripping points from the fault point to the end point, and the set of these tripping points is the optimal set of tripping points;

[0186] The optimal trip point set corresponding to the operating data when historical faults occurred is used to solve the weights of the key features under each fault type. The solution is used to bring the return trip point parameter value into the calculation, thereby obtaining the optimal trip point calculation model.

[0187] That is, the optimal tripping point calculation model includes the following contents: the power grid is regarded as a graphic plane, the fault point is regarded as the starting point, the end of the line is the end point, and the nodes in the middle are the tripping points;

[0188] For each trip point, calculate the parameter value of the trip point under the fault type;

[0189] The parameter value of the trip point is calculated as follows:

[0190] Multiply each key feature selected in (3.3.1) corresponding to the fault type by the weight of the key feature, sum the results, and obtain the parameter value of the trip point;

[0191] Under each fault type, the sum of the weights of all key features is 1;

[0192] The A* algorithm is used to find the minimum path of the sum of the parameter values ​​of the trip points from the fault point to the end point, and the set of these trip points is the optimal set of trip points.

[0193] There are many factors that affect the power supply range. The present invention uses 8 influencing factors that are closely related to the power supply range and intuitively have a linear relationship as independent variables. The correlation analysis between the influencing factors and between each feature and y is performed by writing python code. For example, in a certain power supply area, through calculation, it is obtained that x3, x4 are very weakly correlated with y, x2 is moderately correlated with y, x1, x6, x8 are strongly correlated with y, and x5, x7 are extremely strongly correlated with y. Therefore, except for x3 and x4, other attributes can be used as key attributes for power supply range prediction and analysis, excluding x3 and x4; then Lasso regression is performed.

[0194] (3.3.3) Model verification and optimization

[0195] Provide a visualization interface, supply a large number of samples to the model for training, and adjust the parameter correlation coefficients to optimize the model.

[0196] Step (4), intelligently determine the best trip point:

[0197] When a fault occurs, the optimal trip point calculation model is used to calculate the optimal trip point set, and the corresponding trip point location and the power supply range after the trip point trips are displayed in the topological relationship model;

[0198] By building a visualization application, after collecting power grid fault information, the corresponding single-line diagram is automatically retrieved for analysis, and the model is applied to calculate the optimal tripping point. The optimal tripping point is presented in a visual way on the equipment line diagram, providing a decision-making basis for business personnel to determine the tripping point.

[0199] Step (5), feedback and execution: the business personnel feed back the optimal trip point set to the power grid control system, execute the corresponding tripping operation, and record the tripping event at the same time.

[0200] The present invention can also subsequently continuously optimize and improve the optimal trip point calculation model according to actual operating conditions and feedback data to improve the accuracy and efficiency of the determination.

[0201] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An intelligent method for determining the optimal tripping point based on equipment topology, characterized in that: The steps include: Step (1): Establish a device topology relationship model: The devices in the power grid are abstracted as nodes, and the connection relationship between devices is abstracted as edges, so as to establish the device topology relationship model of the power grid; Step (2): real-time collection of power grid data: Real-time collection of operating data and historical data of equipment in the power grid; Step (3), model construction: Establishing an optimal trip point calculation model based on the power grid topology model obtained in step (1) and the operation data obtained in step (2); Step (4), intelligently determine the best trip point: When a fault occurs, the optimal trip point calculation model is used to calculate the optimal trip point set, and the corresponding trip point location and the power supply range after the trip point trips are displayed in the topological relationship model; Step (5), feedback and execution: The business personnel feed back the optimal trip point set to the power grid control system, execute the corresponding tripping operation, and record the tripping event.

2. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 1 is characterized in that: Step (1) specifically includes the following steps: (1.1) Single-line diagram identification: Get the single-line diagram of the power grid; (1.2) Image preprocessing: Preprocessing the single-line diagram of the acquired power grid; (1.3) Image segmentation: Divide the single-line diagram into multiple areas, each area corresponds to a device or a group of devices; (1.4) Device abstraction: Abstract the devices in the power grid as nodes in the graph; (1.5) Connection relationship identification: Identify the connection relationship between devices; (1.6) Connection relationship abstraction: Abstract the connection relationship between devices as edges in the graph; (1.7) Graph structure construction: Use graph structures to represent devices and their connections.

3. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 2 is characterized in that: In step (1.2), preprocessing includes binarization and denoising.

4. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 2 is characterized in that: In step (1.4), the equipment includes transformers, switches, circuit breakers, feeders, busbars, lines, capacitors and reactors.

5. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 2 is characterized in that: In step (1.5), the connection relationship is identified by method 1 or method 2, as follows: Method 1: Proximity analysis: Analyze the relative positions between devices to determine whether they are connected; Method 2: Path tracking: Starting from one device, trace along a line to another device to determine the connection relationship between them.

6. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 1 is characterized in that: In step (2), the operation data includes the number of transformers, power outage scope, important protection equipment, asynchronous reclosing equipment, faulty equipment, tripping equipment, number of users, and number of topologies.

7. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 1 is characterized in that: The specific method of step (3) is: (3.1) Data preprocessing: Preprocess the operating data when historical faults occur to remove outliers and missing values, and perform normalization; The operation data when the historical fault occurs includes the number of transformers affected, the area of ​​power outage, the number of important protection equipment, the number of asynchronous reclosing equipment, the number of faulty equipment, the number of tripped equipment, the number of users and the number of topologies when the historical fault occurs and after the trip; (3.2) Data classification: Classify the collected data into equipment faults and line faults according to the fault type; (3.3) Optimal trip point calculation model model construction: (3.3.1) Correlation analysis Taking the power supply area as y, analyze the correlation between the eight influencing factors, including the number of transformers x1, the power outage area x2, the number of important protection equipment x3, the number of asynchronous reclosing equipment x4, the number of faulty equipment x5, the number of tripping equipment x6, the number of users x7, and the number of topologies x8, and the power supply area as y; Specifically, for the two fault types, the Pearson correlation coefficient method is used to analyze the correlation between x1~x8 and y, and the Lasso regression is used to remove the collinearity problem between the influencing factors, so as to extract the key features that affect the power supply range for the two fault types; (3.3.2) Construction of optimal trip point calculation model: The power grid is considered as a graphic plane, with the fault point as the starting point, the end of the line as the end point, and the nodes in the middle as the various tripping points; For each trip point, calculate the parameter value of the trip point under the fault type; The parameter value of the trip point is calculated as follows: Multiply each key feature selected in (3.3.1) corresponding to the fault type by the weight of the key feature, sum the results, and obtain the parameter value of the trip point; Under each fault type, the sum of the weights of all key features is 1; The A* algorithm is used to find the minimum path of the sum of the parameter values ​​of the tripping points from the fault point to the end point, and the set of these tripping points is the optimal set of tripping points; The optimal trip point set corresponding to the operating data when historical faults occurred is used to solve the weights of the key features under each fault type. The solution is used to bring the return trip point parameter value into the calculation, thereby obtaining the optimal trip point calculation model.

8. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 7 is characterized in that: In (3.2), the fault types mentioned include line faults and equipment faults.

9. The optimal trip point intelligent determination method based on equipment topology relationship according to claim 7 is characterized in that: In (3.3.1), when using the Pearson correlation coefficient method to analyze the correlation between x1~x8 and y, select the influencing factors with a correlation coefficient greater than 0.5, and then perform Lasso regression.

Images