Fault diagnosis system and method based on power dispatching data network

Abstract

The invention relates to the technical field of electric power data monitoring, in particular to a fault diagnosis system and method based on an electric power dispatching data network, and the method comprises the steps: analyzing the equipment operation state of each network node based on the equipment operation data of each network node in the electric power dispatching data network, mining a fault association rule, and constructing a node fault knowledge base; predicting a high-probability fault node and a fault time period, and determining a node fault degree; constructing a dynamic network topology model of the nodes based on a graph theory, confirming a standby line set, analyzing the power supply stability of standby lines, and generating a line priority list; and switching the fault node into an optimal standby line confirmed by the line priority list, temporarily supplying power, performing real-time monitoring after the standby line is successfully switched, further executing a process of main and standby line switching-standby line no-load detection-resource release, and re-distributing the fault node into the power dispatching data network. And safe and stable operation of a power system is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power data monitoring, and in particular to a fault diagnosis system and method based on an electric power dispatching data network. Background Art

[0002] With the continuous improvement of social science and technology, computers and the Internet are widely used in people's daily work and life, which has significantly improved the level of intelligence and digitalization of the national power grid. The power dispatching data network plays a vital role in ensuring the reliability and security of real-time data communication between dispatching centers, as well as between dispatching centers and substations and power plants.

[0003] The power dispatching data network plays a critical role in data transmission within the power system, and its stability directly impacts the reliable operation of the power system. A network failure can lead to power dispatch failures and even widespread power outages. Traditional fault diagnosis methods rely on manual investigation, which is inefficient and inaccurate, making them inadequate for the rapid fault location and repair required by modern power systems. Furthermore, when a power grid fault occurs and the faulty node is located, network operations personnel are often required to perform repairs, during which time users are unable to access normal electricity.

[0004] Therefore, how to timely predict the fault nodes in the power grid and how to effectively allocate its power backup lines under the fault nodes are urgent problems to be solved. Summary of the Invention

[0005] The purpose of the present invention is to provide a fault diagnosis system and method based on a power dispatching data network to solve the problems raised in the above background technology.

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

[0007] A fault diagnosis method based on a power dispatching data network, the method comprising:

[0008] S100, based on the equipment operation data of each network node in the power dispatching data network, analyzing the equipment operation status of each network node; mapping the equipment operation status, service performance data and configuration association data of each network node into a standardized feature vector, mining fault association rules, and building a node fault knowledge base;

[0009] S200, based on the historical fault records and network load curves of each network node, predict the high-probability fault nodes and fault time periods, further combine the node fault knowledge base to build a fault judgment matrix, and confirm the node fault degree;

[0010] S300: Obtain device information for a faulty node, then construct a dynamic network topology model of the node based on graph theory to identify its backup line set; obtain historical power supply records for each backup line in the backup line set, analyze the power supply stability of the backup line, and establish a comprehensive evaluation model based on the line load rate and physical distance of the backup line to generate a line priority list;

[0011] S400: In the line priority list, the faulty node is switched to the optimal backup line confirmed by the line priority list for temporary power supply. If the switching fails, the suboptimal line is selected. After the backup line is successfully switched, real-time monitoring is performed. After the faulty node is repaired, the system executes the "main-backup line switching-backup line no-load detection-resource release" process to reallocate the faulty node to the power dispatching data network.

[0012] Preferably, S100 includes:

[0013] S101. Acquire equipment operation data of a network node in a power dispatching data network in real time, extract power data values ​​that can reflect the operating status of the network node, and construct a power data curve for the network node with the time node as the horizontal axis and the power data value as the vertical axis;

[0014] Based on the power data values ​​at each time node in the power data curve, with a set interval as one cycle, the power data values ​​in the interval are captured respectively, and the data mean e1 and standard deviation e2 of the power data in the interval are calculated; the power data mean E in the power data curve is obtained, and when α1|Eu×e1|+α2e2 exceeds the threshold for n consecutive cycles, an early warning is triggered and the equipment operation status of a certain network node is marked as abnormal;

[0015] Among them, α1 and α2 represent power anomaly parameters, and u represents the number of cycles in the power data curve;

[0016] S102. All network nodes marked as abnormal equipment operation status are set as fault nodes, and the business performance data and configuration-related data of the fault nodes are obtained respectively. The corresponding power data values, equipment locations, business performance data, and configuration-related data are mapped into standardized feature vectors of the fault nodes. The Apriori algorithm is used to mine fault association rules, identify the abnormal type of the fault node, and build a fault knowledge base for the power dispatching data network.

[0017] Preferably, S200 includes:

[0018] S201. Assume that there are r fault nodes. Based on the fault knowledge base of the power dispatching data network, obtain the historical fault records and load curve data of each fault node, and extract the fault interval sequence and load characteristics of each fault node from them; then, based on the fault interval sequence and load characteristics, use the Cox proportional risk model to describe the failure risk of a fault node at time t as h(t|x)=h0(t)exp(β1x1+β2x2+…+β p x p );

[0019] Where h0(t) represents the benchmark risk function, X=[x1,x2,…,x p ] represents the characteristic vector of a fault node composed of the fault interval sequence and load characteristics, and β represents the regression coefficient;

[0020] Divide a fixed interval into time periods, calculate the mean fault risk within each fixed interval in the fault interval sequence, and then sort each fault node in descending order according to the mean fault risk within the time period, thereby obtaining the fault risk mean sorting sequence of the fault nodes in each time period;

[0021] S202, obtain the fault index system {b1, b2, ..., bm}, where {b1, b2, ..., bm} represents a fault index system consisting of m indicators of impact range index, urgency index and historical risk index, and determine the pairwise comparison matrix through expert scoring

[0022] Among them, a ij Indicates the importance of indicator i relative to indicator j, with a scale ranging from 1 to 9;

[0023] S203, obtain the maximum eigenvalue and corresponding eigenvector of the pairwise comparison matrix A, and obtain the weight vector W=[w1,w2,…,w m ], and further construct the fault judgment matrix according to the fault risk mean and weight vector W of each fault node in a certain time period Thus, the node failure degree value f of the fault node i in a certain time period is obtained ij .

[0024] Preferably, S300 includes:

[0025] S301. Obtain information about the faulty node device with the highest node fault severity value in the current time period, and collect a backup line set of the faulty node based on the dynamic network topology model; obtain g historical power supply records of a backup line in the backup line set, confirm the switching frequency v1 of the backup line in the backup line set, and the average power supply capacity change v2 and the average service switching loss change v3 in the g historical power supply records, thereby obtaining the line stability value of the backup line:

[0026] Where Q represents the total capacity of the backup line, Y represents the construction cost of the backup line, and δ represents the line impact parameter;

[0027] S302: Obtain the line load rate and physical distance of all backup lines in the current time period, and build a comprehensive evaluation model of the fault node with respect to the backup line set based on the line stability value, and generate a line priority list;

[0028] S303 : After the faulty node with the highest fault severity is matched with the line priority list, each faulty node in the fault knowledge base is matched with the line priority list in descending order according to the fault severity value of the node.

[0029] A fault diagnosis system, comprising: a fault diagnosis module, a node analysis module, a line evaluation module and a power call module;

[0030] The fault diagnosis module analyzes the device operation status of each network node based on the device operation data of each network node in the power dispatching data network; maps the device operation status, service performance data and configuration association data of each network node into a standardized feature vector, mines fault association rules, and builds a node fault knowledge base;

[0031] The node analysis module predicts high-probability fault nodes and fault time periods based on the historical fault records and network load curves of each network node, and further builds a fault judgment matrix based on the node fault knowledge base to confirm the degree of node fault;

[0032] The line assessment module obtains device information of a faulty node, constructs a dynamic network topology model of the node based on graph theory, and identifies its backup line set; obtains historical power supply records of each backup line in the backup line set, analyzes the power supply stability of the backup line, and establishes a comprehensive assessment model based on the line load rate and physical distance of the backup line to generate a line priority list;

[0033] The power call module switches the faulty node to the optimal backup line confirmed by the line priority list for temporary power supply. If the switch fails, the suboptimal line is selected. After the backup line is successfully switched, real-time monitoring is carried out. After the faulty node is repaired, the system executes the "main-backup line switching-backup line no-load detection-resource release" process to reallocate the faulty node to the power dispatching data network.

[0034] Preferably, the fault diagnosis module includes: a device status analysis unit and a node fault diagnosis unit;

[0035] The device status analysis unit is used to analyze the device operation status of each network node; the node fault diagnosis unit is used to mine fault association rules and build a node fault knowledge base.

[0036] Preferably, the node analysis module includes: a fault prediction unit and a degree judgment unit;

[0037] The fault prediction unit is used to predict high-probability fault nodes and fault time periods based on the historical fault records of each network node and the network load curve; the degree judgment unit is used to construct a fault judgment matrix in combination with the node fault knowledge base to confirm the node fault degree.

[0038] Preferably, the line evaluation module includes: a backup line acquisition unit, a stability analysis unit and a line evaluation unit;

[0039] The line acquisition unit constructs a dynamic network topology model of the node based on graph theory to confirm its backup line set; the stability analysis unit is used to analyze the power supply stability of the backup line based on the historical power supply records of each backup line; the line evaluation unit is used to establish a comprehensive evaluation model based on the line load rate and physical distance of the backup line to generate a line priority list.

[0040] Preferably, the power calling module includes: a line confirmation unit and a repair switching unit;

[0041] The line confirmation unit is used to switch the fault node to the optimal backup line confirmed by the line priority list; the repair switching unit is used to execute the "main and backup line switching-backup line no-load detection-resource release" process to reallocate the fault node to the power dispatching data network.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] Through a rational system architecture design, an effective fault diagnosis algorithm, and a scientific backup line selection strategy, this invention enables rapid and accurate diagnosis of power dispatch data network faults, timely deployment of appropriate backup lines, and safeguarding the safe and stable operation of the power system. System implementation and verification have demonstrated the high feasibility and practicality of this method, providing strong technical support for the operation and maintenance of power systems. Future research will further optimize the fault diagnosis algorithm to enhance the system's ability to diagnose complex faults. Furthermore, by integrating artificial intelligence and big data technologies, intelligent management and optimized scheduling of backup lines can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0045] Figure 1 This is a flow chart of a fault diagnosis system and method based on a power dispatching data network of the present invention;

[0046] Figure 2 It is a structural diagram of a fault diagnosis system and method based on an electric power dispatching data network of the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] See also Figure 1-Figure 2 , the present invention provides a technical solution:

[0049] Example 1: A fault diagnosis method based on a power dispatching data network, the method comprising:

[0050] S100, acquiring in real time the equipment operation data, service performance data, and configuration-related data of each network node in the power dispatching data network; analyzing the equipment operation status of each network node based on the equipment operation data of each network node in the power dispatching data network; mapping the equipment operation status, service performance data, and configuration-related data of each network node into a standardized feature vector, using the Apriori algorithm to mine fault association rules, and constructing a node fault knowledge base;

[0051] In the power dispatching data network, data collection equipment is deployed at key network nodes, such as substations, power plants, and dispatch centers. These devices include smart sensors and network probes, which are used to collect real-time equipment operating data (such as device temperature, CPU usage, and memory usage), network status data (such as link bandwidth utilization, packet loss rate, and latency), and power business data (such as voltage, current, and power). For example, by installing smart sensors on network devices such as routers and switches, information such as the device's port status and interface traffic can be obtained; network probes can be used to capture network packets and analyze their protocol type, source and destination addresses, and other content.

[0052] Preferably, S100 includes:

[0053] S101. Acquire equipment operation data of a network node in a power dispatching data network in real time, extract power data values ​​that can reflect the operating status of the network node, and construct a power data curve for the network node with the time node as the horizontal axis and the power data value as the vertical axis;

[0054] Based on the power data values ​​at each time node in the power data curve, with a set interval as one cycle, the power data values ​​in the interval are captured respectively, and the data mean e1 and standard deviation e2 of the power data in the interval are calculated; the power data mean E in the power data curve is obtained, and when α1|Eu×e1|+α2e2 exceeds the threshold for n consecutive cycles, an early warning is triggered and the equipment operation status of a certain network node is marked as abnormal;

[0055] Among them, α1 and α2 represent power anomaly parameters, and u represents the number of cycles in the power data curve;

[0056] S102. All network nodes marked as abnormal equipment operation status are set as fault nodes, and the business performance data and configuration-related data of the fault nodes are obtained respectively. The corresponding power data values, equipment locations, business performance data, and configuration-related data are mapped into standardized feature vectors of the fault nodes. The Apriori algorithm is used to mine fault association rules, identify the abnormal type of the fault node, and build a fault knowledge base for the power dispatching data network.

[0057] S200, based on the historical fault records and network load curves of each network node, predict the high-probability fault nodes and fault time periods, further combine the node fault knowledge base to build a fault judgment matrix, and confirm the node fault degree;

[0058] Preferably, S200 includes:

[0059] S201. Assume that there are r fault nodes. Based on the fault knowledge base of the power dispatching data network, obtain the historical fault records and load curve data of each fault node, and extract the fault interval sequence and load characteristics of each fault node from them; then, based on the fault interval sequence and load characteristics, use the Cox proportional risk model to describe the failure risk of a fault node at time t as h(t|x)=h0(t)exp(β1x1+β2x2+…+β p x p );

[0060] Where h0(t) represents the benchmark risk function, X=[x1,x2,…,x p ] represents the characteristic vector of a fault node composed of the fault interval sequence and load characteristics, and β represents the regression coefficient;

[0061] The historical fault records include node ID, fault timestamp, fault type (such as hardware failure, communication interruption, etc.), repair time, etc.; the load curve data includes CPU utilization, bandwidth usage, etc.

[0062] Divide a fixed interval into time periods, calculate the mean fault risk within each fixed interval in the fault interval sequence, and then sort each fault node in descending order according to the mean fault risk within the time period, thereby obtaining the fault risk mean sorting sequence of the fault nodes in each time period;

[0063] S202, obtain the fault index system {b1, b2, ..., bm}, where {b1, b2, ..., bm} represents a fault index system consisting of m indicators of impact range index, urgency index and historical risk index, and determine the pairwise comparison matrix through expert scoring

[0064] Impact indicators include the scope of fault impact and system load fluctuation; urgency indicators include repair difficulty and average repair time; historical risk indicators include the number of faults in the past year and the coefficient of variation of fault intervals;

[0065] Among them, a ij Indicates the importance of indicator i relative to indicator j, with a scale of 1-9; for example, 1 = equally important, 3 = slightly important, and 9 = extremely important;

[0066] S203, obtain the maximum eigenvalue and corresponding eigenvector of the pairwise comparison matrix A, and obtain the weight vector W=[w1,w2,…,w m ], and further construct the fault judgment matrix according to the fault risk mean and weight vector W of each fault node in a certain time period Thus, the node failure degree value f of the fault node i in a certain time period is obtainedij .

[0067] For example, if a switch's failure rate is found to increase during high-temperature periods (2:00 PM to 4:00 PM in July and August), it is more advantageous to pre-migrate part of the node's traffic to a backup line in advance to reduce pressure on the primary line.

[0068] S300: Obtain device information of a faulty node, including device number, physical location, and network logical address, then construct a dynamic network topology model of the node based on graph theory to identify its backup line set; obtain historical power supply records of each backup line in the backup line set, analyze the power supply stability of the backup line, and establish a comprehensive evaluation model based on the line load rate and physical distance of the backup line to generate a line priority list;

[0069] In the dynamic network topology model, nodes are regarded as vertices of the graph, lines are regarded as edges, and each edge is marked with real-time load rate, bandwidth capacity, and transmission delay parameters.

[0070] Preferably, S300 includes:

[0071] S301. Obtain information about the faulty node device with the highest node fault severity value in the current time period, and collect a backup line set of the faulty node based on the dynamic network topology model; obtain g historical power supply records of a backup line in the backup line set, confirm the switching frequency v1 of the backup line in the backup line set, and the average power supply capacity change v2 and the average service switching loss change v3 in the g historical power supply records, thereby obtaining the line stability value of the backup line:

[0072] in, Indicates the resource idle rate, represents the switching cost ratio; Q represents the total capacity of the backup line, Y represents the construction cost of the backup line, and δ represents the line impact parameter;

[0073] S302: Obtain the line load rate and physical distance of all backup lines in the current time period, and build a comprehensive evaluation model of the fault node with respect to the backup line set based on the line stability value, and generate a line priority list;

[0074] S303 : After the faulty node with the highest fault severity is matched with the line priority list, each faulty node in the fault knowledge base is matched with the line priority list in descending order according to the fault severity value of the node.

[0075] S400: In the line priority list, the faulty node is switched to the optimal backup line confirmed by the line priority list for temporary power supply. If the switching fails, the suboptimal line is selected. After the backup line is successfully switched, real-time monitoring is performed. After the faulty node is repaired, the system executes the "main-backup line switching-backup line no-load detection-resource release" process to reallocate the faulty node to the power dispatching data network.

[0076] After the fault is repaired, the system executes the "primary-standby line switching - standby line idle detection - resource release" process, which includes: ① Prioritize switching high-priority services back to the primary line and gradually migrate low-priority traffic; ② Perform health status checks on the idle standby line (such as link connectivity, device temperature, and port bit error rate); ③ Release redundant resources (such as shutting down the power of idle standby equipment and adjusting the bandwidth allocation strategy to the optimal state).

[0077] Example 2: A fault diagnosis system, comprising: a fault diagnosis module, a node analysis module, a line evaluation module, and a power call module;

[0078] The fault diagnosis module analyzes the device operation status of each network node based on the device operation data of each network node in the power dispatching data network; maps the device operation status, service performance data and configuration association data of each network node into a standardized feature vector, mines fault association rules, and builds a node fault knowledge base;

[0079] The node analysis module predicts high-probability fault nodes and fault time periods based on the historical fault records and network load curves of each network node, and further builds a fault judgment matrix based on the node fault knowledge base to confirm the degree of node fault;

[0080] The line assessment module obtains device information of a faulty node, constructs a dynamic network topology model of the node based on graph theory, and identifies its backup line set; obtains historical power supply records of each backup line in the backup line set, analyzes the power supply stability of the backup line, and establishes a comprehensive assessment model based on the line load rate and physical distance of the backup line to generate a line priority list;

[0081] The power call module switches the faulty node to the optimal backup line confirmed by the line priority list for temporary power supply. If the switch fails, the suboptimal line is selected. After the backup line is successfully switched, real-time monitoring is carried out. After the faulty node is repaired, the system executes the "main-backup line switching-backup line no-load detection-resource release" process to reallocate the faulty node to the power dispatching data network.

[0082] Preferably, the fault diagnosis module includes: a device status analysis unit and a node fault diagnosis unit;

[0083] The device status analysis unit is used to analyze the device operation status of each network node; the node fault diagnosis unit is used to mine fault association rules and build a node fault knowledge base.

[0084] Preferably, the node analysis module includes: a fault prediction unit and a degree judgment unit;

[0085] The fault prediction unit is used to predict high-probability fault nodes and fault time periods based on the historical fault records of each network node and the network load curve; the degree judgment unit is used to construct a fault judgment matrix in combination with the node fault knowledge base to confirm the node fault degree.

[0086] Preferably, the line evaluation module includes: a backup line acquisition unit, a stability analysis unit and a line evaluation unit;

[0087] The line acquisition unit constructs a dynamic network topology model of the node based on graph theory to confirm its backup line set; the stability analysis unit is used to analyze the power supply stability of the backup line based on the historical power supply records of each backup line; the line evaluation unit is used to establish a comprehensive evaluation model based on the line load rate and physical distance of the backup line to generate a line priority list.

[0088] Preferably, the power calling module includes: a line confirmation unit and a repair switching unit;

[0089] The line confirmation unit is used to switch the fault node to the optimal backup line confirmed by the line priority list; the repair switching unit is used to execute the "main and backup line switching-backup line no-load detection-resource release" process to reallocate the fault node to the power dispatching data network.

[0090] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A fault diagnosis method based on a power dispatching data network, characterized by: The method comprises: S100, based on the equipment operation data of each network node in the power dispatching data network, analyzing the equipment operation status of each network node; mapping the equipment operation status, service performance data and configuration association data of each network node into a standardized feature vector, mining fault association rules, and building a node fault knowledge base; S200, based on the historical fault records and network load curves of each network node, predict the high-probability fault nodes and fault time periods, further combine the node fault knowledge base to build a fault judgment matrix, and confirm the node fault degree; S300: Obtain device information for a faulty node, then construct a dynamic network topology model of the node based on graph theory to identify its backup line set; obtain historical power supply records for each backup line in the backup line set, analyze the power supply stability of the backup line, and establish a comprehensive evaluation model based on the line load rate and physical distance of the backup line to generate a line priority list; S400: In the line priority list, the faulty node is switched to the optimal backup line confirmed by the line priority list for temporary power supply. If the switching fails, the suboptimal line is selected. After the backup line is successfully switched, real-time monitoring is performed. After the faulty node is repaired, the system executes the "main-backup line switching-backup line no-load detection-resource release" process to reallocate the faulty node to the power dispatching data network.

2. A fault diagnosis method based on a power dispatching data network according to claim 1, characterized in that: The S100 includes: S101. Acquire equipment operation data of a network node in a power dispatching data network in real time, extract power data values ​​that can reflect the operating status of the network node, and construct a power data curve for the network node with the time node as the horizontal axis and the power data value as the vertical axis; Based on the power data values ​​at each time node in the power data curve, with a set interval as one cycle, the power data values ​​in the interval are captured respectively, and the data mean e1 and standard deviation e2 of the power data in the interval are calculated; the power data mean E in the power data curve is obtained, and when α1|Eu×e1|+α2e2 exceeds the threshold for n consecutive cycles, an early warning is triggered and the equipment operation status of a certain network node is marked as abnormal; Among them, α1 and α2 represent power anomaly parameters, and u represents the number of cycles in the power data curve; S102. All network nodes marked as abnormal equipment operation status are set as fault nodes, and the business performance data and configuration-related data of the fault nodes are obtained respectively. The corresponding power data values, equipment locations, business performance data, and configuration-related data are mapped into standardized feature vectors of the fault nodes. The Apriori algorithm is used to mine fault association rules, identify the abnormal type of the fault node, and build a fault knowledge base for the power dispatching data network.

3. A fault diagnosis method based on a power dispatching data network according to claim 1, characterized in that: The S200 includes: S201. Assume that there are r fault nodes. Based on the fault knowledge base of the power dispatching data network, obtain the historical fault records and load curve data of each fault node, and extract the fault interval sequence and load characteristics of each fault node from them; then, based on the fault interval sequence and load characteristics, use the Cox proportional risk model to describe the failure risk of a fault node at time t as h(t|x)=h0(t)exp(β1x1+β2x2+…+β p x p ); Where h0(t) represents the benchmark risk function, X=[x1,x2,…,x p ] represents the characteristic vector of a fault node composed of the fault interval sequence and load characteristics, and β represents the regression coefficient; Divide a fixed interval into time periods, calculate the mean fault risk within each fixed interval in the fault interval sequence, and then sort each fault node in descending order according to the mean fault risk within the time period, thereby obtaining the fault risk mean sorting sequence of the fault nodes in each time period; S202, obtain the fault index system {b1, b2, ..., bm}, where {b1, b2, ..., bm} represents a fault index system consisting of m indicators of impact range index, urgency index and historical risk index, and determine the pairwise comparison matrix through expert scoring Among them, a ij Indicates the importance of indicator i relative to indicator j, with a scale ranging from 1 to 9; S203, obtain the maximum eigenvalue and corresponding eigenvector of the pairwise comparison matrix A, and obtain the weight vector W=[w1,w2,…,w m ], and further construct the fault judgment matrix according to the fault risk mean and weight vector W of each fault node in a certain time period Thus, the node failure degree value f of the fault node i in a certain time period is obtained ij .

4. A fault diagnosis method based on a power dispatching data network according to claim 1, characterized in that: The S300 includes: S301. Obtain information about the faulty node device with the highest node fault severity value in the current time period, and collect a backup line set of the faulty node based on the dynamic network topology model; obtain g historical power supply records of a backup line in the backup line set, confirm the switching frequency v1 of the backup line in the backup line set, and the average power supply capacity change v2 and the average service switching loss change v3 in the g historical power supply records, thereby obtaining the line stability value of the backup line: Where Q represents the total capacity of the backup line, Y represents the construction cost of the backup line, and δ represents the line impact parameter; S302: Obtain the line load rate and physical distance of all backup lines in the current time period, and build a comprehensive evaluation model of the fault node with respect to the backup line set based on the line stability value, and generate a line priority list; S303 : After the faulty node with the highest fault severity is matched with the line priority list, each faulty node in the fault knowledge base is matched with the line priority list in descending order according to the fault severity value of the node.

5. A fault diagnosis system for implementing the fault diagnosis method based on a power dispatching data network according to any one of claims 1 to 4, characterized in that: The system includes: a fault diagnosis module, a node analysis module, a line evaluation module and a power call module; The fault diagnosis module analyzes the device operation status of each network node based on the device operation data of each network node in the power dispatching data network; maps the device operation status, service performance data and configuration association data of each network node into a standardized feature vector, mines fault association rules, and builds a node fault knowledge base; The node analysis module predicts high-probability fault nodes and fault time periods based on the historical fault records and network load curves of each network node, and further builds a fault judgment matrix based on the node fault knowledge base to confirm the degree of node fault; The line assessment module obtains device information of a faulty node, constructs a dynamic network topology model of the node based on graph theory, and identifies its backup line set; obtains historical power supply records of each backup line in the backup line set, analyzes the power supply stability of the backup line, and establishes a comprehensive assessment model based on the line load rate and physical distance of the backup line to generate a line priority list; The power call module switches the faulty node to the optimal backup line confirmed by the line priority list for temporary power supply. If the switch fails, the suboptimal line is selected. After the backup line is successfully switched, real-time monitoring is carried out. After the faulty node is repaired, the system executes the "main-backup line switching-backup line no-load detection-resource release" process to reallocate the faulty node to the power dispatching data network.

6. The fault diagnosis system according to claim 5, wherein: The fault diagnosis module includes: a device status analysis unit and a node fault diagnosis unit; The device status analysis unit is used to analyze the device operation status of each network node; the node fault diagnosis unit is used to mine fault association rules and build a node fault knowledge base.

7. The fault diagnosis system according to claim 5, wherein: The node analysis module includes: a fault prediction unit and a degree judgment unit; The fault prediction unit is used to predict high-probability fault nodes and fault time periods based on the historical fault records of each network node and the network load curve; the degree judgment unit is used to construct a fault judgment matrix in combination with the node fault knowledge base to confirm the node fault degree.

8. The fault diagnosis system according to claim 5, wherein: The line evaluation module includes: a standby line acquisition unit, a stability analysis unit and a line evaluation unit; The line acquisition unit constructs a dynamic network topology model of the node based on graph theory to confirm its backup line set; the stability analysis unit is used to analyze the power supply stability of the backup line based on the historical power supply records of each backup line; the line evaluation unit is used to establish a comprehensive evaluation model based on the line load rate and physical distance of the backup line to generate a line priority list.

9. The fault diagnosis system according to claim 5, wherein: The power call module includes: a line confirmation unit and a repair switching unit; The line confirmation unit is used to switch the fault node to the optimal backup line confirmed by the line priority list; the repair switching unit is used to execute the "main and backup line switching-backup line no-load detection-resource release" process to reallocate the fault node to the power dispatching data network.

Images