Accurate diagnosis and evaluation method and system for distribution network line

Abstract

The invention provides a distribution network line accurate diagnosis and evaluation method and system, and belongs to the technical field of power distribution networks, and the method specifically comprises the steps: obtaining the constitution data of interference risk monitoring devices in different power distribution network regions, and combining the correlation condition between fault risk positions and the interference risk monitoring devices in the power distribution network regions, and determining optimization target areas in the power distribution network area, and determining an evaluation analysis method of the fault diagnosis model according to fault alarm data of the interference risk monitoring devices in different optimization target areas in the alarm event and distribution data of the interference risk monitoring devices in the optimization target areas. And the accuracy of an evaluation analysis result of the identification reliability of the fault diagnosis model is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power distribution networks, and particularly relates to a power distribution network line accurate diagnosis and evaluation method and system. BACKGROUND

[0002] In order to realize fault detection of the line of the power distribution network, in the invention patent application CN202511484491.6 "Low-voltage power distribution network fault positioning method based on electric meter cooperative communication", each intelligent meter generates initial fault information in real time according to monitoring electrical parameters and based on the electrical parameters, and sends the initial fault information to the main processor through the communication topology, thereby improving the fault detection efficiency, the process is simple, and no additional manpower and material resources are needed, but the following defects exist: When performing fault diagnosis, it is often necessary to analyze the alarm signal of the monitoring device to identify the fault position in the power distribution network line, and once the operation state of some monitoring devices is not good, there may be false alarm signals from the monitoring device although the monitored object has not occurred abnormally, so that the accuracy of the diagnosis result may be affected, and therefore how to not perform operation and maintenance processing on the monitoring device with interference risk in the partial area to realize targeted interference risk monitoring and determine the reliability of the diagnosis processing of the fault diagnosis model becomes a technical problem to be solved.

[0003] To solve the above technical problems, the application provides a power distribution network line accurate diagnosis and evaluation method and system. SUMMARY

[0004] To achieve the purpose of the application, the application adopts the following technical solutions: Specifically, the application provides a power distribution network line accurate diagnosis and evaluation method, which specifically comprises: S1, determining the interference data of the alarm signal of the monitoring device when the fault diagnosis model of the power distribution network occurs a fault based on the analysis result of the fault diagnosis model, determining the interference data of the monitoring device in different fault events and the interference situation of the fault diagnosis result of the fault diagnosis model when the interference data of the alarm signal needs to be considered based on the composition situation of the interference data in different fault events, and determining the interference risk monitoring device in the monitoring device; S2, acquiring the composition data of the interference risk monitoring device in different power distribution network areas, and determining the optimization target area in the power distribution network area based on the association situation between the fault risk position in the power distribution network area and the interference risk monitoring device; S3, determining the evaluation analysis method of the fault diagnosis model according to the fault alarm data of the interference risk monitoring device in the alarm event in different optimization target areas and the distribution data of the interference risk monitoring device in the optimization target area.

[0005] The present application has the advantages of: The interference risk monitoring device composition data in different power distribution network regions, the association between the fault risk position in the power distribution network region and the interference risk monitoring device, the determination of the optimization target region in the power distribution network region, taking into account the number of interference risk monitoring devices, and also taking into account the interference probability caused by whether it is in the same power transmission line as the fault risk position, realizing the determination of the optimization target region from the perspective of interference probability, thereby laying the foundation for further determining the accuracy of the fault identification result of the fault diagnosis model when the interference risk monitoring device sends a false alarm signal.

[0006] According to the fault alarm data of the interference risk monitoring device in the different optimization target regions in the alarm event and the distribution data of the interference risk monitoring device in the optimization target region, the determination of the evaluation analysis method of the fault diagnosis model is performed, which takes into account the newly added interference of the interference risk monitoring device in the optimization target region in the alarm event, and also takes into account the influence of the number of optimization target regions with interference risk monitoring devices on the operation stability of the entire power distribution network, realizing the determination of the evaluation analysis method of the fault diagnosis model from the perspective of influence and interference, i.e. whether to consider the interference data of the interference risk monitoring device, laying the foundation for as soon as possible to perform operation and maintenance processing of the interference risk monitoring device and improve the operation stability of the power distribution network.

[0007] Further, the interference data of the alarm signal of the monitoring device is determined according to the data that the monitoring object of the monitoring device has not failed, but has sent a false alarm signal in a fault event.

[0008] Further, determining that the interference data of the alarm signal needs to be considered, specifically includes: Determine the fault event with interference data according to the composition of the interference data in different fault events in the power distribution network. According to the fault event with interference data, determine whether to consider the interference data of the alarm signal.

[0009] Further, the method for determining the optimization target region in the power distribution network region is: Determine the number of interference risk monitoring devices in the power distribution network region according to the composition data of the interference risk monitoring devices in the power distribution network region. Determine the fault risk position in the power distribution network based on the fault events in history at different positions in the power distribution network region, and determine the distance between the fault risk position and the interference risk monitoring device based on the association between the fault risk position and the interference risk monitoring device. According to the number of interference risk monitoring devices in the power distribution network region and the distance between the fault risk position and the interference risk monitoring devices, it is determined whether the power distribution network region is an optimization target region.

[0010] In another aspect, the present application provides a computer system comprising a memory and a processor connected in communication, and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to perform the power distribution line accurate diagnosis and evaluation method.

[0011] Other features and advantages will be set forth in the descriptions that follow, and in part will be apparent from the description, or can be learned by practice of the application. The purposes and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims.

[0012] In order to make the above objectives, features and advantages of the present application more apparent, the following will describe a preferred embodiment in detail, with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0013] The above and other features and advantages of the present application will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings in which:

[0014] Figure 1 is a flowchart of a power distribution line accurate diagnosis and evaluation method; Figure 2 is a flowchart of determining interference data that needs to consider alarm signals; Figure 3 is a flowchart of a method of determining interference risk monitoring devices in monitoring devices. DETAILED DESCRIPTION

[0015] Example embodiments now will be described more fully hereinafter with reference to the accompanying drawings. Example embodiments, however, can be implemented in many different forms and should not be construed as limited to the implementations set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example embodiments to those skilled in the art. Like reference numerals refer to like elements throughout the specification.

[0016] The terms "one", "a", "an", "said", and "the" are used to indicate the existence of one or more than one element / portion / etc.; the terms "including" and "having" are used to indicate an open-ended inclusion of elements / portion / etc. in the description of a process, method, or composition of matter.

[0017] Example 1 To solve the above problems, according to one aspect of the present application, as shown in Figure 1 A distribution network line accurate diagnosis and evaluation method is provided, which specifically comprises: S1 determines the interference data of the alarm signal of the monitoring device when the fault occurs in the distribution network based on the analysis result of the fault diagnosis model of the distribution network, determines the interference data of the monitoring device in different fault events based on the composition of the interference data in different fault events, and determines the interference risk monitoring device in the monitoring device when the interference data of the alarm signal needs to be considered, and the interference situation of the fault diagnosis result of the fault diagnosis model; S2 obtains the composition data of the interference risk monitoring device in different distribution network regions, and determines the optimization target region in the distribution network region in combination with the association between the fault risk position in the distribution network region and the interference risk monitoring device; S3 determines the evaluation analysis method of the fault diagnosis model according to the fault alarm data of the interference risk monitoring device in the alarm event in different optimization target regions and the distribution data of the interference risk monitoring device in the optimization target region.

[0018] Further, the interference data of the alarm signal of the monitoring device is determined according to the data that the monitoring object of the monitoring device does not occur fault, but error occurs alarm signal in the fault event.

[0019] Specifically, the interference data refers to the "false alarm" data in this context. That is, the monitoring object (such as switch cabinet, line) of the monitoring device itself does not occur fault, but the device incorrectly issues an alarm signal during a certain fault event.

[0020] It can be understood that the fault diagnosis model is a system based on artificial intelligence or rules, and the input is the alarm signal uploaded by various monitoring devices in the power grid, and the output is the judgment of the fault position and type.

[0021] Interference situation evaluation: quantifying the influence degree of false alarm signal on the final judgment result of the fault diagnosis model.

[0022] Specifically, as shown in Figure 2 The determination of the interference data of the alarm signal specifically comprises: S11 determines the fault events with interference data based on the composition of the interference data in different fault events in the distribution network; Specifically, suppose we analyze 100 recorded fault events of a certain distribution network in the past year. Focus on the alarm behavior of 10 common monitoring devices (such as overcurrent protection, differential protection, partial discharge monitoring, etc.).

[0023] In the above step, 100 fault events are analyzed retrospectively to check the alarm records of all monitoring devices in each event.

[0024] Criteria: If a monitoring device issues an alarm, but the post-event verification proves that the equipment or line section it is responsible for monitoring is completely free of problems, then this alarm is marked as a "false alarm", and the fault event is marked as a "fault event with interference data".

[0025] Simulation results: Among the 100 fault events, it is found that in 8 fault events, at least one monitoring device has a false alarm.

[0026] For example: Fault event #23: a short circuit occurs on a 10kV line. The line main protection correctly acts, but a set of partial discharge monitoring devices on the adjacent line also incorrectly report a "partial discharge exceeds" alarm, Fault event #47: a transformer fault. The transformer differential protection correctly acts, but an unrelated switch cabinet temperature monitoring device incorrectly reports an "over-temperature" alarm.

[0027] S12 determines whether to consider the interference data of the alarm signal according to the fault event with interference data.

[0028] Specifically, the fault event with interference data is the fault event of the monitoring device with a false alarm.

[0029] It can be understood that when the number of fault events with interference data does not meet the requirement, that is, is greater than the preset threshold, the number of fault events of the monitoring device with a false alarm is relatively large, and therefore, in order to realize the assessment of the influence degree of the diagnosis result of the fault event on the monitoring device with a false alarm, it is determined that the interference data of the alarm signal needs to be considered, thereby realizing the assessment of the recognition reliability of the fault diagnosis model under different interference data, and further laying a foundation for the determination of the targeted operation and maintenance strategy.

[0030] Specifically, whether to consider the interference data is determined by counting the number of "fault events with interference data" and comparing it with the preset threshold.

[0031] Calculation and judgment: the number of fault events with interference data = 8 times, the preset threshold = 5 times, judgment: 8>5? Yes, therefore, it is determined that the interference data of the alarm signal needs to be considered.

[0032] If the false alarm event only occurs sporadically (such as less than 5 times), it can be regarded as an occasional phenomenon, and the influence on the overall diagnosis system can be ignored.

[0033] But when the number exceeds the threshold, it indicates that false alarms have become a systemic problem, not just an isolated case. It can frequently pollute the data source input to the fault diagnosis model. Without considering these interference data, the evaluation of the reliability of the model is "paper tiger", which cannot reflect its real performance in the actual noisy environment.

[0034] Specifically, the interference situation of the fault diagnosis result of the fault diagnosis model is determined according to the consistency degree of the fault diagnosis result when the alarm signal of the monitoring device with false alarm exists and the alarm signal of the monitoring device without false alarm exists.

[0035] Specifically, as shown in Figure 3 The method for determining the interference risk monitoring device in the monitoring device is: The core goal of this embodiment is to accurately identify those devices that not only frequently misreport, but also substantially interfere with fault diagnosis results from numerous monitoring devices, and mark them as "interference risk monitoring devices" to facilitate priority evaluation and analysis of the interference impact of the fault diagnosis model.

[0036] S21 determines the fault event in which the false alarm signal of the monitoring device exists with the interference data of the monitoring device in different fault events, and takes it as a matching fault event; In the above step, the matching fault event is determined. The matching fault event refers to an event in which a specific monitoring device incorrectly issues an alarm signal in an event of a fault in a power distribution network. That is, the device monitored by the device itself has no problem, but it misreports.

[0037] The system analyzes 200 fault events and checks the alarm behavior of each device in each event.

[0038] Device P: false alarm occurred in 8 fault events. The number of matching fault events = 8, device Q: false alarm occurred in 4 fault events. The number of matching fault events = 4, device S: false alarm occurred in 4 fault events. The number of matching fault events = 4.

[0039] This step is the starting point of problem discovery. It filters from the frequency level of the behavior of a device, quantifies the number of times the device "helps the wrong way at the critical moment (when the fault occurs)", which is the basic index for evaluating its reliability, and provides a data pool for subsequent in-depth analysis of its influence.

[0040] S22 determines the deviation of the fault diagnosis result of the matching fault event in the absence of the false alarm signal of the monitoring device from the fault diagnosis result in the presence of the false alarm signal of the monitoring device based on the fault diagnosis results of the fault diagnosis model in different matching fault events, and determines the interference fault event in the matching fault event by using the deviation. In the above step of determining the interference fault event, the interference fault event is an event in the "matching fault event" of a certain device, which causes the final output result (such as fault location, type) of the fault diagnosis model to deviate from the correct result after excluding the false alarm signal.

[0041] Action: For each matching fault event found in S21, a comparison experiment analysis of "with or without interference" is performed. Device P: Among the 8 false alarms, 2 caused the diagnosis model to output errors (such as misjudging the fault location). The number of interference fault events = 2, Device Q: Among the 4 false alarms, 3 caused the diagnosis model to output errors. The number of interference fault events = 3, Device S: Among the 4 false alarms, 1 caused the diagnosis model to output errors. The number of interference fault events = 1.

[0042] This step is the key to the hazard assessment. It filters from the result impact level, distinguishing between "harmless noise" and "harmful interference". A device may often misreport, but if the intelligent system can easily identify and ignore it, its harm is limited. This step ensures that we focus our attention on devices that can really destroy the accuracy of decision-making.

[0043] S23 determines whether the monitoring device is an interference risk monitoring device based on the matching fault event data and the interference fault event in the matching fault event.

[0044] It can be understood that determining whether the monitoring device is an interference risk monitoring device based on the matching fault event data and the interference fault event in the matching fault event specifically includes: S231 determines the number of matching fault events of the monitoring device based on the matching fault event data, judges whether the number of matching fault events of the monitoring device is greater than a preset fault event number threshold, if yes, determines that the monitoring device is an interference risk monitoring device, if no, proceeds to the next step. In the above step, based on the preliminary screening of the false alarm absolute number, this step directly uses the number of matching fault events for judgment.

[0045] Judgment and decision: Device P: number (8) > threshold (5)? Yes → directly determine as an interference risk monitoring device.

[0046] This is a high-efficiency filter. For devices with extremely frequent false positives (such as device P), whether their individual false positives have an impact or not, they themselves have become a huge "data noise source", severely reducing the quality of the entire data pool. For such devices, there is no need for complex calculations, and they are directly listed as high-risk and processed, in line with the "80 / 20 principle" of operation and management.

[0047] S232 obtains the interference fault events of the monitoring device, determines whether the number of interference fault events of the monitoring device is greater than a preset interference event number threshold, if yes, proceeds to the next step, and if no, determines that the monitoring device does not belong to an interference risk monitoring device; In the above step, based on the screening of the absolute number of harmful false positives, this step uses the number of interference fault events for judgment.

[0048] Judgment and decision: Device Q: number (3) > threshold (2)? Yes → enter S233.

[0049] Device S: number (1) > threshold (2)? No → determined as not belonging to an interference risk monitoring device.

[0050] This step checks the absolute number of times that cause substantial harm for devices that do not pass S231. If a device has a small total number of false positives but a large number of times that cause incorrect diagnoses (such as device Q), it is still a major hidden danger and needs further analysis. Conversely, like device S, both the false positives and the number of times that cause impact are small, and it can be safely excluded from the risk list.

[0051] S233 determines an interference risk factor of the matching fault events based on a proportion of the number of interference fault events in the matching fault events, determines whether the interference risk factor of the matching fault events is greater than a preset interference risk factor threshold, if no, proceeds to the next step, and if yes, determines that the monitoring device belongs to an interference risk monitoring device. In the above step, based on the screening of the false positive "toxicity" intensity, the interference risk factor: the calculation formula is the number of interference fault events / the number of matching fault events. It measures the proportion of a device's false alarm signals that will eventually evolve into interference to diagnosis. The higher this ratio, the stronger the false positive "toxicity" of the device and the greater the harm.

[0052] Calculation and judgment: Device Q: interference risk factor = 3 / 4 = 0.75, judgment: 0.75 > threshold (0.4)? Yes → determined as an interference risk monitoring device.

[0053] This step is the accurate positioning of the degree of harm. It reveals the intrinsic characteristics of the device. There is a 75% probability that the false alarm of device Q will mislead the system, which is a very high "failure rate". This means that the signal output by the device is of very low reliability, and once it alarms, it is difficult for the operation and maintenance personnel to believe it, and the diagnostic model is also easily biased by it. Such "high toxicity" devices must be marked.

[0054] S234 determines the interference frequency factor of the monitoring device based on the number of matching fault events of the monitoring device and in combination with a preset proportion factor, determines the interference risk coefficient based on the average of the interference risk factor and the interference frequency factor, and determines whether the monitoring device belongs to the interference risk monitoring device based on the interference risk coefficient.

[0055] In the above steps, the interference frequency factor is based on the final decision of the comprehensive risk coefficient: the number of matching fault events x the preset proportion factor. This is a weighted value of false alarm frequency, so that frequent false alarm behavior in a large power grid (many fault events, large proportion factor) can be given a higher weight.

[0056] The interference risk coefficient is: (interference frequency factor + interference risk factor) / 2. This is a final indicator that combines "behavior frequency" and "result toxicity" to make a final decision on devices in a critical state.

[0057] Scenario deduction (hypothetical): Suppose there is a device T, whose data is at the critical point of all thresholds, the number of matching fault events = 5 (equal to the S231 threshold), the number of interference fault events = 2 (equal to the S232 threshold), and the interference risk factor = 2 / 5 = 0.4 (equal to the S233 threshold). At this time, S231, S232, and S233 cannot make a decision, so S234 is entered.

[0058] Calculation and judgment: interference frequency factor = 5 x 0.1 = 0.5, interference risk coefficient = (0.5 + 0.4) / 2 = 0.45, judgment: 0.45 > preset risk coefficient threshold (0.3)? Yes → finally determine that device T is an interference risk monitoring device.

[0059] This step is the safety net and optimizer of the decision system. It prevents the missed judgment that may occur under simple threshold judgment. For devices such as device T, which have "crossed the line" in all indicators, simple rules may let them pass. However, after comprehensive calculation, the risk coefficient is high, indicating that the comprehensive risk cannot be ignored. The above steps ensure the comprehensiveness and fairness of risk assessment, making the decision more delicate and scientific.

[0060] Through this set of clear and progressive evaluation system, we can: quickly capture obvious problems with the device (such as device P), accurate identification of although not frequent but the device (such as device Q) is huge harm, effectively exclude the impact of the device (such as device S) is very small, prudent decision in critical state of the device (such as device T).

[0061] The final output of the "interference risk monitoring device list" provides a clear priority guide for the operation and maintenance team, realizes the change of operation and maintenance resources from "extensive management" to "precise allocation", and fundamentally improves the accuracy and reliability of power distribution network fault diagnosis.

[0062] It can be understood that the preset proportion factor is determined according to the number of fault events in the power distribution network, wherein the more the number of fault events, the larger the preset proportion factor, and the specific value range is between 0 and 1.

[0063] Specifically, determining whether the monitoring device belongs to an interference risk monitoring device based on the interference risk coefficient, specifically comprising: When the interference risk coefficient is greater than a preset risk coefficient threshold, it is determined that the monitoring device belongs to an interference risk monitoring device.

[0064] Specifically, the composition data of the interference risk monitoring device in the power distribution network region is determined according to the distribution position of the monitoring device of the same type as the interference risk monitoring device in the power distribution network region.

[0065] Specifically, the method for determining the optimization target region in the power distribution network region is: S31 determines the number of interference risk monitoring devices in the power distribution network region based on the composition data of the interference risk monitoring devices in the power distribution network region. In the above steps, the number of regional risk devices is counted, and the composition data of the interference risk monitoring device: refers to the list and number of all (current and potential) interference risk monitoring devices in the region determined based on the distribution of the same type device.

[0066] Step goal: get the total amount of interference risk monitoring devices in the region, action and data: count the number of interference risk monitoring devices = 4.

[0067] This is a preliminary screening of risk. From a macro point of view, grasp the basic reliability of the monitoring system in the region. If a region is full of unreliable monitoring devices, the efficiency of interference risk verification in the region is higher.

[0068] S32, determine a fault risk position in the power distribution network based on the fault events in history at different positions in the power distribution network region, and determine a distance between the fault risk position and the interference risk monitoring device based on an association between the fault risk position and the interference risk monitoring device; Specifically, the fault risk position is identified and the distance is calculated. The fault risk position is a geographic position with a number of historical faults exceeding a preset threshold, which is a physical weak point of the power grid. The distance refers to an electrical distance or a topological connection relationship, rather than a pure geographic distance in meters. That is, it is determined whether a risk device and a fault point are on the same line or on a line with a close electrical connection.

[0069] Step goal: find the physical weak points in the region and analyze their association with unreliable monitoring devices.

[0070] Action and data: identify the number of fault risk positions = 2 (F1, F2).

[0071] Analyze the association: F1 is on the same line L1 as the model X partial discharge instrument, and the distance is short. F2 is on the same line L2 as the model X partial discharge instrument, and the distance is short.

[0072] This is a risk association analysis. The "monitoring risk" is associated with the "physical risk" and paired. Only when the unreliable device is responsible for monitoring the equipment with a high failure probability, the risk is truly formed. This step will abstract the global risk into a "risk pair".

[0073] S33, determine whether the power distribution network region is an optimization target region according to the number of interference risk monitoring devices in the power distribution network region and the distance between the fault risk position and the interference risk monitoring device.

[0074] It should be noted that the fault risk position in the power distribution network is a position with a number of historical fault events greater than a preset event number threshold.

[0075] It can be understood that, according to the number of interference risk monitoring devices in the power distribution network region and the distance between the fault risk position and the interference risk monitoring device, whether the power distribution network region is an optimization target region is determined, specifically including: S331, obtain the number of interference risk monitoring devices in the power distribution network region, determine whether the number of interference risk monitoring devices is greater than a preset monitoring device number threshold, if yes, determine that the power distribution network region is an optimization target region, if not, go to the next step; In the above step, based on the preliminary screening of the absolute number of risk devices, the step aims to quickly identify and lock the area of "risk device flooding", and determine that the number of risk devices, i.e. the number of Model X localizer, is 4, which is not greater than the threshold value (5), and enter S332.

[0076] Specifically, the above step belongs to a high-efficiency filter. For monitoring areas with poor basic quality (risk devices > 5), there is no need for complex analysis, and they are directly listed as the highest priority, which meets the "80 / 20 rule" in management.

[0077] S332 determines whether there is an interference risk monitoring device in the power distribution network area. If yes, it enters the next step. If no, it is determined that the power distribution network area does not belong to the optimization target area. In the above step, the presence of risk devices is checked, and the step aims to confirm whether there is indeed a risk source in the area that needs attention.

[0078] Determination: number of risk devices (4) > 0? Yes, decision: enter S333, which belongs to a prerequisite check. Ensure that the subsequent analysis has a practical object and avoid wasting analysis resources in risk-free areas.

[0079] S333 obtains the number of fault risk positions in the power distribution network area, determines whether the number of fault risk positions in the power distribution network area is greater than a preset risk position number threshold, and if yes, determines that the power distribution network area belongs to the optimization target area, and if no, enters the next step. In the above step, based on the screening of the number of fault risk positions, the step aims to identify areas where faults are very frequent by themselves, and determine whether the number of fault risk positions (2) is greater than the threshold value (3). No, decision: enter S334.

[0080] Inherent risk assessment. For areas where faults occur frequently by themselves, it is generally possible to monitor the interference of interference risk monitoring devices. Such areas should be prioritized for optimization and identification, i.e. not to perform maintenance and operation on interference risk monitoring devices, so as to determine the interference of fault diagnosis results of the fault diagnosis model under the interference of risk monitoring devices.

[0081] S334 determines the power distribution line that simultaneously has a fault risk position and an interference risk monitoring device based on the distance between the fault risk position and the interference risk monitoring device, and takes it as a target power distribution line. Based on the number of fault risk positions in different target power distribution lines, it determines whether the power distribution network area is an optimization target area.

[0082] It can be understood that, based on the number of fault risk positions in different target power distribution lines, it is determined whether the power distribution network area is an optimization target area, which specifically includes: S3341 determines whether the number of target power distribution lines is greater than the preset target line number threshold. If yes, it determines that the power distribution network area belongs to the optimization target area. If no, it proceeds to the next step. Specifically, in the above steps, based on the screening of the number of lines with overlapping risks, the target distribution line is: a line that simultaneously possesses both a "fault risk location" and an "interference risk monitoring device." It is a physical unit of overlapping risks.

[0083] Step objective: Assess the extent of risk overlap from the line dimension. Line L1: Has F1 and Model X device → is the target line. Line L2: Has F2 and Model X device → is the target line. Line L3: No fault risk location → is not the target line. Number of standard power lines = 2.

[0084] Judgment: 2 is not greater than the threshold (2), proceed to S3342.

[0085] Risk distribution breadth assessment. If there are multiple (>2) risk-overlapping lines, it indicates a high probability of interference. Therefore, the distribution network area can be identified as the optimization target area, eliminating the need for operation and maintenance of interference risk monitoring devices and identifying the degree of impact.

[0086] S3342 determines the weight coefficient of different target power distribution lines based on the number of fault risk locations in different target power distribution lines. When the sum of the weight coefficients of different target power distribution lines is greater than the preset weight coefficient threshold, the power distribution network area is determined to belong to the optimization target area.

[0087] In the above steps, the final decision is based on the weighted risk distribution along the target line. The weighting coefficient is determined by the number of fault risk locations on the target line. The greater the number, the higher the weight. This is used to quantify the "density" or "severity" of risks within the line. The weighting coefficient = number of fault risk locations × 1.5. The objective of this step is to assess the severity of the risk distribution and make a final decision.

[0088] Actions and Calculations: Line L1 weight = 1 × 1.5 = 1.5, Line L2 weight = 1 × 1.5 = 1.5, Total weight coefficient = 1.5 + 1.5 = 3.0 Judgment: The total weight coefficient (3.0) is not greater than the threshold (4.0)? Therefore, it is determined that "Chengxi Industrial Zone" does not belong to the optimization target area.

[0089] Risk severity actuarial. This is the most refined assessment. It not only looks at how many risk lines, but also considers the severity of each line risk. The total weight coefficient does not reach the critical point, indicating that the probability of alarm events occurring is not high, so targeted interference risk monitoring device operation and maintenance processing can be carried out, thereby ensuring the stability of the power distribution network operation.

[0090] It should be noted that the weight coefficient of the target power transmission line is determined according to the number of fault risk positions in the target power distribution line, wherein the more the number of fault risk positions in the target power distribution line, the greater the weight coefficient of the target power transmission line.

[0091] Specifically, the method for determining the evaluation and analysis method of the fault diagnosis model is: In this step, the dynamic decision makes the evaluation strategy of the fault diagnosis model. The core choice is: when evaluating the performance of the model, should the false alarm data of these interference risk devices be used as test input (considering interference) to test their anti-interference performance; or should the devices be maintained before evaluation, and the core diagnosis ability should be evaluated under pure data (without considering interference).

[0092] The basic logic: This is a decision about the allocation of "test" and "operation and maintenance" resources. When false alarm risk is widespread and active, prioritize "testing" to optimize the model; when the risk is concentrated but static, prioritize "operation and maintenance" to ensure safety.

[0093] S41 determines the alarm events in which the interference risk monitoring device exists false alarm based on the fault alarm data of the interference risk monitoring device in different optimization target areas in alarm events, and takes them as new alarm events; In the above step, the new alarm event: in the recent monitoring period (such as one month), the false alarm event newly generated by the device marked as "interference risk monitoring device". This reflects the "current activity" of its interference behavior.

[0094] S42 determines the number of interference risk monitoring devices in the optimization target area based on the distribution data of the interference risk monitoring devices in the optimization target area. In the above step, the latest and key data input is collected for decision-making.

[0095] From the monitoring log, filter out the newly generated new alarm events of the interference risk monitoring device in each optimization target area in the recent period, and count the total number of interference risk monitoring devices in each optimization target area.

[0096] This is the basis of the decision. Without accurate and timely data, all analysis is built on air. This step ensures that the decision is based on facts, not speculation.

[0097] Table 1 initial data S43, according to the newly added alarm event of the interference risk monitoring device and the number of interference risk monitoring devices in different optimization target areas, determine the evaluation analysis method of the fault diagnosis model.

[0098] It can be understood that, according to the newly added alarm event of the interference risk monitoring device and the number of interference risk monitoring devices in different optimization target areas, the evaluation analysis method of the fault diagnosis model is determined, which specifically includes: S431, the number of optimization target areas is obtained, and it is judged whether the number of optimization target areas is less than a preset target area number threshold. If yes, even if the interference risk monitoring device is not operated and maintained, the influence degree on the operation stability of the overall power distribution network is not high, so it is determined that the evaluation analysis method of the fault diagnosis model is to consider the interference of all interference risk monitoring devices. If not, go to the next step; In the above step, based on the global preliminary screening of the number of optimization target areas, the "global universality" of the interference risk is evaluated.

[0099] Judgment: optimization target area number (4) < preset threshold (3)? No, decision: enter S432.

[0100] If the number of risk areas is small (<3), it means that the problem is local and has limited impact on the overall power grid. At this time, there is enough time to use these "real interference" to test the anti-interference ability (robustness) of the model comprehensively. Since the risk is universal (4>3), the failure to timely operate and maintain may lead to poor operation reliability of the power distribution network, so more careful evaluation is needed.

[0101] S432, the number of optimization target areas with the interference risk monitoring device is obtained, and it is judged whether the number of optimization target areas with the interference risk monitoring device is less than a preset target area number threshold. If yes, even if the interference risk monitoring device is not operated and maintained, the influence degree on the operation stability of the overall power distribution network is not high, so it is determined that the evaluation analysis method of the fault diagnosis model is to consider the interference of the interference risk monitoring device. If not, go to the next step; Specifically, based on the screening of the number of optimization target areas containing the interference risk monitoring device, it is confirmed that the influence degree of the interference risk monitoring device not being operated and maintained on the operation stability of the entire power distribution network.

[0102] If the number of optimization target regions of the model X device is not less than the preset threshold (2), it means that there are many optimization target regions of the model X device, and further in-depth analysis is needed.

[0103] S433 determines whether there are new alarm events of the interference risk monitoring device based on the new alarm events of the interference risk monitoring device. If yes, it proceeds to the next step. If no, because there are many optimization target regions of the interference risk monitoring device, in order to ensure the operation stability of the power distribution network, it is determined that the evaluation and analysis method of the fault diagnosis model does not need to consider the interference of the interference risk monitoring device, that is, the operation and maintenance processing of the interference risk monitoring device is performed. Specifically, based on the emergency screening of the risk device activity, it is determined whether the risk is in the "silent but potential" state.

[0104] It is determined whether all the interference risk monitoring devices do not have new alarm events. If no (regions B and D are very active), it is determined that S434 is entered.

[0105] This is a "safety first" check. If all high-risk devices are "silent" (no new false alarms) in the current cycle, it means there is no immediate threat. However, since the risk area is widespread, from the long-term safety point of view, the most stable strategy is to directly perform operation and maintenance processing to create a "pure" data environment for model evaluation and prioritize system reliability. Since there are active devices, further analysis is needed.

[0106] S434 determines the monitoring matching factor based on the proportion of the number of interference risk monitoring devices of the interference risk monitoring device with new alarm events in the optimization target region in the optimization target region. Based on the monitoring matching factor in different optimization target regions, the evaluation and analysis method of the fault diagnosis model in the interference risk monitoring device is determined.

[0107] Further, based on the monitoring matching factor in different optimization target regions, the evaluation and analysis method of the fault diagnosis model in the interference risk monitoring device is determined, specifically including: Based on the sum of the monitoring matching factors in different optimization target regions, the monitoring matching coefficient is determined. It is determined whether the monitoring matching coefficient is greater than the preset matching coefficient threshold. If yes, the monitoring reliability of the interference risk of the interference risk monitoring device is high, and it is determined that the evaluation and analysis method of the fault diagnosis model needs to consider the interference of the interference risk monitoring device. If no, it is determined that the evaluation and analysis method of the fault diagnosis model does not need to consider the interference of the interference risk monitoring device, that is, the operation and maintenance processing of the interference risk monitoring device is performed.

[0108] In the above step, the matching factor is monitored: in a certain optimization target area, the proportion of the number of newly added alarm event interference risk monitoring devices to the total number of all interference risk monitoring devices in the area. It quantifies the "active proportion" of high-risk devices in the area. The monitoring matching coefficient is monitored: the sum of the "monitoring matching factors" of all optimization target areas. It quantifies the "overall activity level" of high-risk device interference behavior in the entire power distribution network from a global perspective.

[0109] In the above step, based on the final decision of the global risk activity coefficient, the "overall activity level" of the risk is quantified, and the final decision is made.

[0110] Calculate the monitoring matching coefficient = Area A (0.4) + Area B (0.875) + Area C (0) + Area D (0.833) = 2.108, judge: monitoring matching coefficient (2.108) > preset matching coefficient threshold (1.5)? Yes, determine that the evaluation and analysis method of the fault diagnosis model is "the interference of the interference risk monitoring device needs to be considered".

[0111] This is the core decision logic. The monitoring matching coefficient (2.108) is much higher than the threshold, indicating that the interference behavior is not only universal, but also very active and intense.

[0112] In this case, large-scale operation and maintenance processing is costly and time-consuming. On the contrary, this provides a very valuable "high noise, high pressure" real test environment. The optimal strategy is to use this environment to "stress test" and "anti-interference evaluation" of the fault diagnosis model, expose its weaknesses in bad data, and provide the most direct basis for subsequent model optimization. If the model can maintain good performance in this environment, its reliability will be the strongest.

[0113] Embodiment 2 On the other hand, the present application provides a computer system comprising a memory and a processor connected in communication, and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to perform the power distribution line accurate diagnosis and evaluation method described above.

[0114] Each of the embodiments in the specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0115] The above-described embodiments of the application have special structure and can achieve the desired results. Other embodiments can have different structures and achieve the same results. The purpose of the above-described embodiments is to illustrate the principles of the application and not to limit the scope of the application. The scope of the application is defined by the claims and their equivalents. Other embodiments are within the scope of the claims.

[0116] The above description is merely illustrative of the embodiments of the present application and is not intended to limit the scope of the present application. Various modifications can be made by those skilled in the art based upon the teachings disclosed herein. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present application shall fall within the scope of the claims of the present application.

Claims

1. A method for precise diagnosis and evaluation of a network configuration line, characterized by, Specifically comprising: With the analysis result of the fault diagnosis model of the power distribution network, the interference data of the alarm signal of the monitoring device when the power distribution network fails is determined, based on the composition of the interference data in different fault events, it is determined that the interference data of the alarm signal needs to be considered, the interference data of the monitoring device in different fault events, and the interference situation of the fault diagnosis result of the fault diagnosis model, and the interference risk monitoring device in the monitoring device is determined; Obtain the composition data of the interference risk monitoring device in different power distribution network areas, and determine the optimization target area in the power distribution network area in combination with the association between the fault risk position in the power distribution network area and the interference risk monitoring device. According to the fault alarm data of the interference risk monitoring device in the different optimization target areas in the alarm event and the distribution data of the interference risk monitoring device in the optimization target area, the determination of the evaluation analysis method of the fault diagnosis model is carried out.

2. The network wiring line accurate diagnosis and evaluation method according to claim 1, wherein The interference data of the alarm signal of the monitoring device is determined according to the data that the monitoring object of the monitoring device does not fail, but the alarm signal is sent incorrectly in the fault event.

3. The network wiring line accurate diagnosis and evaluation method according to claim 1, wherein Determine that the interference data of the alarm signal needs to be considered, specifically comprising: With the composition of the interference data in different fault events in the power distribution network, the fault events with interference data are determined. According to the fault events with interference data, it is determined whether the interference data of the alarm signal needs to be considered.

4. The network wiring line accurate diagnosis and evaluation method according to claim 3, wherein The fault event with interference data is the fault event of the monitoring device with false alarm.

5. The network wiring line accurate diagnosis and evaluation method according to claim 3, wherein When the number of fault events with interference data does not meet the requirements, it is determined that the interference data of the alarm signal needs to be considered.

6. The network wiring line accurate diagnosis and evaluation method according to claim 1, wherein The interference situation of the fault diagnosis result of the fault diagnosis model is determined according to the consistency degree of the fault diagnosis result when the alarm signal of the monitoring device with false alarm and the alarm signal of the monitoring device without false alarm.

7. The network wiring line accurate diagnosis and evaluation method according to claim 1, wherein The determination method of the interference risk monitoring device in the monitoring device is: With the interference data of the monitoring device in different fault events, the fault event with false alarm signal of the monitoring device is determined, and it is used as a matching fault event; Based on the fault diagnosis result of the fault diagnosis model in different matching fault events, the deviation situation of the fault diagnosis result of the matching fault event when the monitoring device with false alarm signal is not present and the fault diagnosis result when the monitoring device with false alarm signal is present is determined, and the deviation situation is used to determine the interference fault event in the matching fault event; With the matching fault event data and the interference fault event in the matching fault event, it is determined whether the monitoring device is an interference risk monitoring device.

8. The network wiring line accurate diagnosis and evaluation method according to claim 7, wherein With the matching fault event data and the interference fault event in the matching fault event, it is determined whether the monitoring device is an interference risk monitoring device, specifically comprising: Based on the matching fault event data of the monitoring device, the number of matching fault events of the monitoring device is determined, and when the number of matching fault events of the monitoring device is greater than the preset fault event number threshold, it is determined that the monitoring device is an interference risk monitoring device.

9. The network wiring line accurate diagnosis and evaluation method according to claim 1, wherein The determined method for evaluating and analyzing the fault diagnosis model is: The interference risk monitoring device in different optimization target areas monitors the fault alarm data in the alarm event, determines the alarm event of the interference risk monitoring device existing false alarm, and takes it as a new alarm event; Based on the distribution data of the interference risk monitoring device in the optimization target area, the number of the interference risk monitoring device in the optimization target area is determined; According to the new alarm event of the interference risk monitoring device and the number of the interference risk monitoring device in different optimization target areas, the evaluation and analysis method of the fault diagnosis model is determined.

10. A computer system comprising: The memory and processor connected in communication, and the computer program stored on the memory and capable of running on the processor, characterized in that the processor executes the computer program to perform the precise diagnosis and evaluation method of the network line of any one of claims 1-9.

Images