Self-adaptive intelligent studying and judging method and system for ground fault of distribution network

Abstract

The invention relates to a distribution network ground fault self-adaptive intelligent research and judgment method and system, and relates to the technical field of distribution network ground fault judgment. By monitoring the three-phase voltage of the distribution network, when the phase voltage is out of limit, a grounding fault is judged and research and judgment are started; synchronously executing at least two grounding fault research and judgment algorithms based on different electrical principles; according to a preset research and judgment algorithm priority strategy which can be dynamically adjusted, sorting the algorithm output results, and generating a comprehensive switch-off sequence table; processing a fault according to the sequence table, and comparing an actual grounding line with a research and judgment first result to generate post evaluation data; and finally, dynamically adjusting the priority strategy based on post evaluation data so as to optimize subsequent research and judgment. Through a closed-loop adaptive optimization mechanism, continuous self-improvement of a research and judgment strategy is realized, the defect of poor adaptability of a fixed fusion rule is overcome, and long-term accuracy and reliability of ground fault research and judgment are remarkably improved.

Description

Technical Field

[0001] This invention relates to an adaptive intelligent assessment method and system for distribution network ground faults, belonging to the technical field of distribution network ground fault assessment. Background Technology

[0002] Single-phase ground faults (commonly known as ground failures) occur frequently in power distribution networks, and quickly and accurately locating the faulty line is crucial to ensuring power supply reliability. With the popularization of distribution automation technology, the master station system can acquire multi-dimensional data such as zero-sequence current, phase current, and intelligent terminal signals, which has led to the development of comprehensive judgment methods based on multiple principles to overcome the shortcomings of single methods in terms of adaptability.

[0003] Existing comprehensive judgment methods mostly employ a step-by-step, serial execution of multiple line selection algorithms, ultimately generating a line-pulling sequence table through a fixed and complex set of fusion rules. For example, Chinese invention patent application CN114019300B discloses a comprehensive judgment method for distribution network ground faults based on multi-source signals, including the following steps: Step S1: Perform line selection and location based on complete switch grounding alarm; Step S2: Perform line selection and location based on zero-sequence current amplitude comparison method; Step S3: Use phase current line selection method to determine the ground fault line; Step S4: Use IQ line selection method to sort the ground fault lines; Step S5: Use line selection device to determine the ground fault line; Step S6: Use the reclosing of outgoing switch as a line selection signal of line selection device to participate in the line-pulling sequence ranking; Step S7: Perform line-pulling sequence fusion. The aforementioned technical solutions combine the characteristics of various ground fault signals and improve the master station's ability to locate ground faults by mining historical operational data. This enables timely and rapid location of ground fault sections, shortens the duration of ground faults, improves fault handling efficiency, and reduces grid losses. However, the sequential execution of such technical solutions increases the total judgment time, which may delay fault handling. Furthermore, the core fusion rules of these technical solutions are often based on preset static weights and complex logic (such as calculating the number of support numbers and distinguishing between fusion and independent strategies). The rule design heavily relies on expert experience and is fixed once set, making it impossible to self-optimize based on actual application effects. When the grid structure, grounding method, or line parameters change, the original fixed fusion rules are no longer applicable, leading to a decrease in judgment accuracy and a lack of long-term evolution capability of the system.

[0004] Therefore, there is an urgent need for an adaptive and self-optimizing intelligent method for judging distribution network grounding faults, in order to overcome the limitations of fixed integration rules and improve the accuracy and reliability of long-term system operation. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes an adaptive intelligent assessment method and system for distribution network ground faults.

[0006] The technical solution of the present invention is as follows: On the one hand, the present invention provides an adaptive intelligent assessment method for distribution network ground faults, the method comprising: The distribution automation master station monitors the three-phase voltage of the distribution network in real time. When the phase voltage exceeds the limit, the distribution automation master station simultaneously starts at least two ground fault line judgment algorithms based on different electrical principles or information sources, and obtains the candidate fault lines corresponding to each algorithm as the judgment results. Based on the preset priority strategy for the judgment algorithms, the judgment results output by each judgment algorithm are sorted to generate a comprehensive road ranking table. Fault handling is carried out according to the comprehensive fault sequence table, and after the fault handling is completed, the actual grounded fault line is compared with the first judgment result of the comprehensive fault sequence table to generate post-event evaluation data. Based on the post-evaluation data, the priority strategy of the judgment algorithm is dynamically adjusted, and the adjusted priority strategy is used for the next judgment.

[0007] Preferably, the judgment algorithm includes at least two of the following: zero-sequence current amplitude comparison method based on zero-sequence current amplitude comparison, phase current method based on phase current change, complete switch method based on the result of local detection signal transmission of complete switch, line selection device method based on signal of dedicated line selection device, and centralized judgment method that integrates multiple electrical quantities for centralized calculation.

[0008] Preferably, according to a preset priority strategy for the analysis algorithms, the analysis results output by each analysis algorithm are sorted to generate a comprehensive ranking table. Specifically: Each judgment algorithm is independently configured with a priority, an enable / disable status flag, and a flag indicating whether the judgment result is included in the ranking; Filter out the judgment results corresponding to the judgment algorithms that are in a disabled state or whose judgment results are not included in the sorting. Sort the judgment results of the judgment algorithms that are in an enabled state and whose judgment results are included in the sorting from high to low priority to obtain the comprehensive route order table. Among them, for the judgment results of different judgment algorithms that output the same route, the judgment result with the highest priority is used as the sorting basis of the same route in the comprehensive route order table.

[0009] Preferably, the configuration basis of the preset judgment algorithm priority strategy includes at least one of the following: historical statistics on the adaptability of different judgment algorithms to the current distribution network operating conditions, statistics on the judgment accuracy of different judgment algorithms in historical faults, and dispatcher's empirical preference settings.

[0010] Preferably, the method further includes: Fault handling is performed according to the comprehensive fault sequence table. If a fault handling completion signal is not received within a preset time, the distribution automation master station automatically initiates a secondary analysis. The new results generated by the secondary analysis are pushed and updated to the comprehensive fault sequence table in a visually prominent manner, but the original order generated based on the results of the first analysis is not changed.

[0011] Preferably, the post-event evaluation data includes fault event identifiers, results of each judgment method, first judgment result, actual grounding line, and an identifier indicating whether the judgment is correct or not. If the first test line pulled by the dispatcher according to the integrated line pull sequence table is the actual grounding line, then the first judgment result is determined to be correct.

[0012] Preferably, the rule for determining that a phase voltage exceeds the limit is that in the three-phase voltage of the distribution network, there is a phase voltage that is higher than the set upper voltage limit threshold, and at the same time there is another phase voltage that is lower than the set lower voltage limit threshold.

[0013] On the other hand, the present invention also provides an adaptive intelligent assessment system for distribution network ground faults, the system comprising: The multi-algorithm parallel startup module is used to monitor the three-phase voltage of the distribution network in real time. When the phase voltage exceeds the limit, at least two ground fault line judgment algorithms based on different electrical principles or information sources are started simultaneously to obtain the candidate fault lines corresponding to each algorithm as the judgment results. The intelligent result sorting module is used to sort the judgment results output by each judgment algorithm according to the preset judgment algorithm priority strategy and generate a comprehensive ranking table. The performance evaluation module is used to handle faults according to the integrated circuit sequence table, and after the fault handling is completed, compare the actual grounded fault line with the first judgment result of the integrated circuit sequence table to generate post-event evaluation data. The strategy self-learning optimization module is used to dynamically adjust the priority strategy of the judgment algorithm based on the post-evaluation data, and to use the adjusted priority strategy of the judgment algorithm for the next judgment.

[0014] In another aspect, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the adaptive intelligent judgment method for distribution network ground faults as described in any one of the present invention.

[0015] In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the program is executed by a processor, it implements an adaptive intelligent judgment method for distribution network ground faults as described in any one of the present invention.

[0016] The present invention has the following beneficial effects: 1. This invention provides an adaptive intelligent assessment method and system for distribution network ground faults. By simultaneously launching at least two assessment algorithms based on different electrical principles and performing parallel calculations, it fully utilizes multi-dimensional fault characteristics and significantly improves the efficiency and comprehensiveness of initial fault assessment. 2. This invention provides an adaptive intelligent judgment method and system for distribution network ground loss faults. By intelligently integrating and sorting the results of multiple algorithms according to a priority strategy that can be configured differently and dynamically adjusted, the decision-making logic is clear and flexible, improving the accuracy of the comprehensive judgment results and the credibility of the dispatcher's decision. 3. This invention provides an adaptive intelligent judgment method and system for distribution network ground faults. By establishing a post-evaluation closed loop with the correctness of the first judgment as the core, and dynamically adjusting the priority strategy of the judgment algorithm based on the evaluation data, the system has the ability to continuously learn from historical experience and optimize itself, fundamentally improving the system's long-term adaptability to different power grid operating conditions and the ability to continuously improve the overall judgment efficiency. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0020] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0022] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0023] Example 1: like Figure 1 As shown in the figure, this embodiment provides an adaptive intelligent judgment method for distribution network ground faults, the method including: S1. The distribution automation master station continuously collects and monitors the three-phase phase voltage of key nodes in the distribution network (such as the 10kV bus of the substation) at a frequency of no less than once per second through the data acquisition system. When the phase voltage exceeds the limit, the distribution automation master station simultaneously starts at least two ground fault line judgment algorithms based on different electrical principles or information sources, and obtains the candidate fault lines corresponding to each algorithm as the judgment results. In this embodiment, a threshold for determining if the phase voltage exceeds the limit is preset. Specifically, this embodiment uses the rated phase voltage... Set an upper voltage threshold as a reference. It is 1.15 lower threshold It is 0.85 The detection of phase voltage exceeding the limit specifically means that one phase voltage is higher than the set upper voltage threshold, and at the same time another phase voltage is lower than the set lower voltage threshold, and this continues for more than a predetermined stability criterion period (e.g., 5 power frequency cycles, i.e. 0.1 seconds). Once it is determined that the phase voltage exceeds the limit, the judgment algorithm is triggered in parallel and synchronously. Furthermore, the judgment algorithm includes at least two of the following: zero-sequence current amplitude comparison method based on zero-sequence current amplitude comparison, phase current method based on phase current change, complete switch method based on the result of local detection signal transmission of complete switch, line selection device method based on dedicated line selection device signal, and centralized judgment method that integrates multiple electrical quantities for centralized calculation. To clarify the point, the specific implementation logic of the several judgment algorithms included in this embodiment is now explained in detail: The zero-sequence current amplitude comparison method immediately reads the effective value of the zero-sequence current of all outgoing lines in the distribution network after the fault starts, and calculates the average value of the zero-sequence current of all lines. ; to convert the zero-sequence current of each line Compare with the average value of the zero-sequence current to filter out those that meet the requirements. ( Lines with a reliability coefficient ranging from 2.5 to 3.5 are considered as a set of suspicious lines; the line with the largest zero-sequence current amplitude is selected from the set of suspicious lines and output as the judgment result of the algorithm. The phase current method focuses on the sudden change in phase current caused by the fault. For each line, the effective values ​​of the three-phase currents A, B, and C in the first cycle after the fault and the first cycle before the fault are calculated respectively. The maximum value of the change among the three phases is taken. If the maximum value of the change of a line exceeds the preset percentage threshold of its rated current (30% in this embodiment), the line is marked as suspicious. The line with the largest value of the change that exceeds the limit is taken as the judgment result of the algorithm. The complete set of switch method relies on intelligent terminals (such as FTU / DTU) deployed on the distribution network. The master station listens to the local detection alarm signal of ground fault sent by the intelligent terminal. When a terminal at a certain line section switch reports a ground fault signal, the power supply feeder to which the switch belongs (i.e. the entire line from the substation bus to the switch) is immediately identified as a candidate fault line. If multiple switches alarm on the same feeder, the switch that alarms first or is closest to the end of the line shall be used. The centralized judgment method synchronously collects information such as the amplitude and phase of zero-sequence voltage and zero-sequence current, harmonic components (such as the 5th harmonic), and transient waveforms of each line at the time of the fault. It performs comprehensive calculation and voting through the built-in composite criterion model (in this embodiment, it first performs group amplitude and phase comparison, then harmonic direction comparison, and finally calculates transient energy distribution). The composite criterion model outputs the line with the highest confidence score as the judgment result, and gives the corresponding confidence score.

[0024] S2. Based on the preset priority strategy for the analysis algorithms, sort the analysis results output by each analysis algorithm to generate a comprehensive ranking table. Specifically: Each judgment algorithm is independently configured with a priority, an enable / disable status flag, and a flag indicating whether the judgment result is included in the ranking. The preset judgment method priority strategy is shown in Table 1: Table 1 Priority Strategy Table for Judgment Methods

[0025] The priority in the table is a positive integer. The smaller the value, the higher the priority and the greater the weight in the sorting. Priority 1 is the highest. The "Enable" flag controls whether the judgment algorithm is executed when a fault occurs. If not, the algorithm will not be started in S1 and will be disabled. The "Participate in Sorting" flag controls whether the judgment result of the algorithm enters the subsequent sorting process. If not, the result of the algorithm is only used for recording and display and does not participate in the generation of the road order table. In this embodiment, the judgment results corresponding to the judgment algorithms that are in a disabled state or whose judgment results are not included in the sorting are filtered out. According to the priority of the judgment algorithms that are in an enabled state and whose judgment results are included in the sorting, the judgment results of their output are sorted from high to low (i.e., the values ​​are from small to large) to obtain a comprehensive route order table. Among them, for the judgment results of different judgment algorithms that output the same route, the judgment result with the highest priority (i.e., the smallest value) is used as the sorting basis of the same route in the comprehensive route order table. To clarify, the specific generation logic is explained in detail below: Collect the judgment results of all judgment algorithms that were started in step S1 and whose sorting indicator is "yes". Each judgment result contains (method ID, candidate line ID). Iterate through all the judgment results and merge the results that point to the same line. For each line judged by different algorithms, find the one with the smallest priority value (i.e. the highest priority) among all the algorithms that reported the line and use that value as the representative priority of the line. For example, if line L1 is reported by both the zero-sequence current amplitude method (priority 1) and the phase current method (priority 3), then the representative priority of line L1 is 1. All unique candidate routes are sorted in ascending order according to their representative priority values. If two routes have the same representative priority, they can be sorted in a secondary order according to the number of methods reported for that route (i.e., the number of votes), with the route with more votes ranked first. If the number of votes is also the same, the final order can be determined by the route number or other rules. Based on the above sorting, an ordered list of routes is generated, namely the comprehensive route order table. The first route in the comprehensive route order table is the highest priority route recommended after comprehensive evaluation, the second route is the next highest priority route, and so on. Furthermore, the configuration basis of the preset judgment algorithm priority strategy includes at least one of the following: historical statistics on the adaptability of different judgment algorithms to the current distribution network operating conditions, statistics on the judgment accuracy of different judgment algorithms in historical faults, and dispatcher's empirical preference settings, wherein: The accuracy statistics of the judgment are specifically calculated by determining the accuracy rate (number of correct judgments / total number of participations) of each judgment algorithm in the historical fault data. When optimizing the preset judgment algorithm priority strategy, the priority of methods with high historical accuracy is increased. The adaptive historical statistics of operating conditions are specifically matched with different priority strategy templates based on the current power grid operating mode (such as neutral point grounding mode), weather conditions and time period. For example, in the neutral point grounding mode via arc suppression coil, a strategy to increase the priority of harmonic method or transient method is automatically activated; in thunderstorm weather, the priority of transient analysis method is increased. The dispatcher's experience preference setting is specifically set by senior dispatchers manually in the strategy table as a basic strategy based on the characteristics of the regional power grid and long-term operating experience.

[0026] S3. Perform fault handling according to the comprehensive fault sequence table, and after the fault handling is completed, compare the actual grounded fault line with the first judgment result of the comprehensive fault sequence table to generate post-event evaluation data, wherein: S31. The specific fault handling according to the comprehensive road sequence table is as follows: The dispatcher receives the comprehensive line pull sequence table pushed by the system on the human-machine interface. According to the order in the table, the dispatcher performs remote test pull operations on the lines in sequence. When a line is tested, the main station detects that the ground fault characteristic signal (such as bus zero-sequence voltage alarm) disappears and the line load current becomes zero. Then the line is confirmed to be the actual ground fault line. Preferably, the confirmation information fed back by the on-site operation and maintenance personnel can also be used as the final criterion. S32. The post-event evaluation data includes fault event identifiers, results of each judgment method, the first judgment result, the actual grounding line, and an identifier indicating whether the judgment was correct or not, wherein: The correctness indicator for the first-order assessment is a Boolean value (True / False). The determination rule is to compare whether the actual grounding line is consistent with the line ranked first in the comprehensive line-pulling sequence table (i.e., the first-order assessment result). If they are consistent, it is marked as correct. In addition, to encourage dispatchers to trust the system recommendation, as long as the dispatcher's first trial line is equal to the actual grounding line, regardless of whether the line is ranked first in the sequence table, it is considered that this round of system assessment effectively assisted the decision-making when calculating system effectiveness. However, the correctness indicator for the first-order assessment is still strictly determined according to the above rules. S33. Furthermore, fault handling is performed according to the comprehensive fault sequence table. If a fault handling completion signal is not received within a preset time, the distribution automation master station automatically initiates a secondary analysis. The new results generated by the secondary analysis are pushed and updated to the comprehensive fault sequence table in a visually prominent manner, but the original order generated based on the results of the first analysis is not changed. In this embodiment: The trigger condition for the secondary assessment is that after the primary assessment is completed and the circuit breaker sequence table is pushed, the system starts a timer (set to a preset time of 2 minutes). If no line test operation signal is received within this time and the fault state has not disappeared, the secondary assessment is automatically triggered. The secondary assessment does not rerun all methods, but focuses on the steps of analyzing the steady-state or new transient characteristics after the fault develops (such as recalculating the steady-state zero-sequence component and analyzing the subsequent arc grounding transient process). Some assessment methods (such as the complete set of switch methods) may update the results due to the terminal sending signals again. The new results generated by the second assessment (which may be a new route or a reconfirmation of an existing route) will be prominently displayed on the original route sequence table interface with a highlighted background color (such as yellow) or a flashing animation. Preferably, the second assessment results do not change the original sorting order generated according to the strategy at the time of the first assessment, and the original list remains unchanged. The second results are displayed as additional prompts floating or marked next to the relevant routes, so that the judgment object for the first assessment result in the post-event evaluation is always the sequence table generated in the first time, while providing dispatchers with more reference information.

[0027] S4. Based on the post-evaluation data, dynamically adjust the priority strategy of the judgment algorithm, and use the adjusted priority strategy for the next judgment. In this embodiment: S41. The dynamically adjusted triggering mechanism includes, but is not limited to, three modes: periodic triggering, event triggering, and manual triggering. These three modes can be used individually or in combination, wherein: The periodic triggering refers to the system automatically starting a round of strategy optimization analysis every fixed period (such as monthly or quarterly); the event triggering refers to the automatic triggering of optimization when the number of accumulated post-evaluation records reaches a preset threshold (such as 100 times); the manual triggering refers to the operation and maintenance personnel being able to manually initiate strategy optimization commands at any time through the system interface. S42. After triggering optimization, the system performs in-depth analysis of the stored post-event evaluation data to assess the performance and overall effectiveness of each judgment method under the current priority strategy. The analysis includes at least individual method performance statistics, operational adaptability analysis, and overall strategy effectiveness evaluation, among which: The individual method performance statistics are specifically calculated for each judgment algorithm (such as the zero-sequence current amplitude comparison method and the phase current method) to calculate its hit rate (i.e., the proportion of times its output candidate line matches the actual grounding line to the total number of participations) and first-place contribution rate (i.e., the number of times its output result becomes the first line in the comprehensive road sequence table). The operating condition adaptability analysis specifically involves associating grid operating condition information (such as neutral grounding method, weather, whether it is high-resistance grounding, etc.) when a fault occurs, and analyzing the differences in the performance of various judgment methods under different operating conditions. For example, the analysis shows that the hit rate of the phase current method decreases significantly under thunderstorm weather. The overall strategy effectiveness evaluation specifically involves statistically analyzing the accuracy rate of the first-order judgment in the comprehensive road order table under the current priority strategy. S43. After generating the analysis report through the above analysis, adjust the priority strategy table for the judgment methods, including: S431. Generate a visual analysis report that clearly shows information such as the recent hit rate of the zero-sequence current amplitude comparison method reaching 92%, which is the primary reliable algorithm, and the frequent false alarms of the complete set of switch method in the arc suppression coil grounding mode. Based on the analysis report and their own experience, maintenance personnel can manually modify the priority value of specific methods or whether they are enabled in the strategy configuration interface. S432. Pre-set optimization algorithm, automatically adjust strategy according to clear performance indicators, such as statistically analyzing the hit rate of each judgment algorithm in the recent N (e.g., the most recent 50) effective faults, sorting them from high to low hit rate, and automatically setting the priority of the method ranked first to the highest (e.g., set to 1), and successively lowering the priority of other methods. S433. Automatically identify the current or predicted dominant power grid operating conditions (such as high-resistance grounding mode). When a specific mode is identified, automatically retrieve the priority strategy template (a complete set of priority and activation status configuration) that has performed best in history under this mode from the historical database and apply it to realize intelligent scenario switching of the strategy. The new priority strategy, after the above adjustments, will be immediately applied to the assessment of the next loss of ground fault.

[0028] Example 2: This embodiment provides an adaptive intelligent assessment system for distribution network ground faults, the system comprising: The multi-algorithm parallel startup module is used to monitor the three-phase voltage of the distribution network in real time. When the phase voltage exceeds the limit, at least two ground fault line judgment algorithms based on different electrical principles or information sources are started simultaneously to obtain the candidate fault lines corresponding to each algorithm as the judgment results. The intelligent result sorting module is used to sort the judgment results output by each judgment algorithm according to the preset judgment algorithm priority strategy and generate a comprehensive ranking table. The performance evaluation module is used to handle faults according to the integrated circuit sequence table, and after the fault handling is completed, compare the actual grounded fault line with the first judgment result of the integrated circuit sequence table to generate post-event evaluation data. The strategy self-learning optimization module is used to dynamically adjust the priority strategy of the judgment algorithm based on the post-evaluation data, and to use the adjusted priority strategy for the next judgment. Example 3: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements an adaptive intelligent judgment method for distribution network ground faults as described in any one of the embodiments.

[0029] Example 4: This embodiment provides a computer-readable storage medium storing a computer program thereon, characterized in that, when the program is executed by a processor, it implements an adaptive intelligent judgment method for distribution network ground faults as described in any one of the embodiments.

[0030] It is worth noting that the system, electronic device, and computer-readable storage medium described in this invention are based on the same inventive concept as the method described in Embodiment 1, and will not be repeated here.

[0031] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0032] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0033] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0034] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0035] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. An adaptive intelligent judgment method for distribution network ground faults, characterized in that, The method includes: The distribution automation master station monitors the three-phase voltage of the distribution network in real time. When the phase voltage exceeds the limit, the distribution automation master station simultaneously starts at least two ground fault line judgment algorithms based on different electrical principles or information sources, and obtains the candidate fault lines corresponding to each algorithm as the judgment results. Based on the preset priority strategy for the judgment algorithms, the judgment results output by each judgment algorithm are sorted to generate a comprehensive road ranking table. Fault handling is carried out according to the comprehensive fault sequence table, and after the fault handling is completed, the actual grounded fault line is compared with the first judgment result of the comprehensive fault sequence table to generate post-event evaluation data. Based on the post-evaluation data, the priority strategy of the judgment algorithm is dynamically adjusted, and the adjusted priority strategy is used for the next judgment.

2. The adaptive intelligent judgment method for distribution network ground faults according to claim 1, characterized in that, The judgment algorithm includes at least two of the following: zero-sequence current amplitude comparison method based on zero-sequence current amplitude comparison, phase current method based on phase current change, complete switch method based on the results of local detection signals transmitted from the complete switch, line selection device method based on signals from a dedicated line selection device, and centralized judgment method that integrates multiple electrical quantities for centralized calculation.

3. The adaptive intelligent judgment method for distribution network ground faults according to claim 1, characterized in that, Based on the preset priority strategy for the analysis algorithms, the analysis results output by each algorithm are sorted to generate a comprehensive ranking table. Specifically: Each judgment algorithm is independently configured with a priority, an enable / disable status flag, and a flag indicating whether the judgment result is included in the ranking; Filter out the judgment results corresponding to the judgment algorithms that are in a disabled state or whose judgment results are not included in the sorting. Sort the judgment results of the judgment algorithms that are in an enabled state and whose judgment results are included in the sorting from high to low priority to obtain the comprehensive route order table. Among them, for the judgment results of different judgment algorithms that output the same route, the judgment result with the highest priority is used as the sorting basis of the same route in the comprehensive route order table.

4. The adaptive intelligent judgment method for distribution network ground faults according to claim 3, characterized in that, The configuration basis of the preset judgment algorithm priority strategy includes at least one of the following: historical statistics on the adaptability of different judgment algorithms to the current distribution network operating conditions, statistics on the judgment accuracy of different judgment algorithms in historical faults, and dispatcher's empirical preference settings.

5. The adaptive intelligent judgment method for distribution network ground faults according to claim 1, characterized in that, The method further includes: Fault handling is performed according to the comprehensive fault sequence table. If a fault handling completion signal is not received within a preset time, the distribution automation master station automatically initiates a secondary analysis. The new results generated by the secondary analysis are pushed and updated to the comprehensive fault sequence table in a visually prominent manner, but the original order generated based on the results of the first analysis is not changed.

6. The adaptive intelligent judgment method for distribution network ground faults according to any one of claims 1 or 5, characterized in that, The post-event evaluation data includes fault event identifiers, results of each judgment method, first judgment result, actual grounding line, and an identifier indicating whether the judgment is correct or not. If the first test line pulled by the dispatcher according to the integrated line pull sequence table is the actual grounding line, then the first judgment result is determined to be correct.

7. The adaptive intelligent judgment method for distribution network ground faults according to claim 1, characterized in that, The rule for determining if a phase voltage exceeds the limit is that in the three-phase voltage of the distribution network, there is one phase voltage that is higher than the set upper voltage limit threshold, and at the same time there is another phase voltage that is lower than the set lower voltage limit threshold.

8. An adaptive intelligent judgment system for distribution network ground faults, characterized in that, The system includes: The multi-algorithm parallel startup module is used to monitor the three-phase voltage of the distribution network in real time. When the phase voltage exceeds the limit, at least two ground fault line judgment algorithms based on different electrical principles or information sources are started simultaneously to obtain the candidate fault lines corresponding to each algorithm as the judgment results. The intelligent result sorting module is used to sort the judgment results output by each judgment algorithm according to the preset judgment algorithm priority strategy and generate a comprehensive ranking table. The performance evaluation module is used to handle faults according to the integrated circuit sequence table, and after the fault handling is completed, compare the actual grounded fault line with the first judgment result of the integrated circuit sequence table to generate post-event evaluation data. The strategy self-learning optimization module is used to dynamically adjust the priority strategy of the judgment algorithm based on the post-evaluation data, and to use the adjusted priority strategy of the judgment algorithm for the next judgment.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements an adaptive intelligent judgment method for distribution network ground faults as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements an adaptive intelligent judgment method for distribution network ground faults as described in any one of claims 1 to 7.

Images