A power distribution network fault active research method and terminal

Abstract

The application discloses a power distribution network fault active research and judgment method and a terminal. A corresponding target decision table is constructed according to the topological structure of a power distribution network. The target decision table contains each condition attribute and a corresponding result attribute. When a fault feature of a fault position in the power distribution network is acquired, the fault feature is matched in the target decision table first to determine a first candidate set of fault elements. Then, a fault diagnosis prediction model corresponding to each candidate fault element is acquired based on the first candidate set. The fault diagnosis prediction model is used to determine a fault credibility corresponding to each candidate fault element. Finally, a fuzzy integral value corresponding to each candidate fault element is determined based on a fault credibility set. The element of the power distribution network where the fault occurs is determined based on the fuzzy integral value, and the active research and judgment of the fault of the power distribution network are realized accurately and quickly.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution network, and particularly relates to a power distribution network fault active research and judgment method and a terminal. BACKGROUND

[0002] The low-voltage power distribution network is the terminal link of the power distribution system and directly serves users. According to statistics, the low-voltage power distribution network users account for more than 90% of the entire power users. Therefore, the reliable operation of the low-voltage power distribution network and the active and timely research and judgment of the fault and the rapid repair are crucial to improving the power supply reliability and the high-quality service level.

[0003] However, for a long time, the power distribution network fault research and judgment mainly focuses on the 10kV power distribution network, and the research and judgment range is from the substation outlet switch to the 10KV branch line. According to the power distribution network fault statistics and analysis in recent years, the low-voltage power distribution network (0.4KV) fault accounts for more than 92% of the entire power distribution network fault. Developing the low-voltage power distribution network fault research and judgment is of great significance to improving the power distribution network state control and operation and maintenance management penetration.

[0004] The traditional power distribution network fault diagnosis prediction model based on a wireless sensor has the problems of inaccurate fault position judgment and large time consumption in the diagnosis process of the power distribution network data in a big data environment. SUMMARY

[0005] The present application solves the technical problem of providing a power distribution network fault active research and judgment method and a terminal, which can accurately and quickly actively research and judge the fault of the power distribution network.

[0006] In order to solve the above technical problems, a technical solution adopted by the present application is as follows:

[0007] A power distribution network fault active research and judgment method, comprising the steps of:

[0008] S1, acquiring the topology structure information of the power distribution network, constructing a corresponding target decision table according to the topology structure information, each record in the target decision table comprising a condition attribute and a corresponding result attribute;

[0009] S2, acquiring the fault characteristics of the fault occurrence position in the power distribution network, determining the matched condition attribute from the target decision table according to the fault characteristics, and determining the corresponding target result attribute according to the matched condition attribute;

[0010] S3, determining a first candidate set of fault elements according to the target result attribute, and acquiring a corresponding fault diagnosis prediction model based on each candidate fault element in the first candidate set;

[0011] S4, determine the fault credibility corresponding to each candidate fault element according to the fault diagnosis prediction model, and form a fault credibility set;

[0012] S5, determine the fuzzy integral value corresponding to each candidate fault element based on the fault credibility set, and determine the element of the power distribution network that has a fault according to the fuzzy integral value.

[0013] In order to solve the above technical problems, another technical solution adopted by the present application is:

[0014] A power distribution network fault active research terminal, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0015] S1, obtain the topology structure information of the power distribution network, and construct a corresponding target decision table according to the topology structure information, each record in the target decision table comprising a condition attribute and a corresponding result attribute;

[0016] S2, obtain the fault characteristics of the fault occurrence position in the power distribution network, determine the matched condition attribute from the target decision table according to the fault characteristics, and determine the corresponding target result attribute according to the matched condition attribute;

[0017] S3, determine a first candidate set of fault elements according to the target result attribute, and obtain a corresponding fault diagnosis prediction model based on each candidate fault element in the first candidate set;

[0018] S4, determine the fault credibility corresponding to each candidate fault element according to the fault diagnosis prediction model, and form a fault credibility set;

[0019] S5, determine the fuzzy integral value corresponding to each candidate fault element based on the fault credibility set, and determine the element of the power distribution network that has a fault according to the fuzzy integral value.

[0020] The application has the beneficial effects that: a corresponding target decision table is constructed according to the topological structure of the power distribution network, the target decision table contains each condition attribute and the corresponding result attribute, when the fault feature of the fault position in the power distribution network is obtained, the fault feature is matched in the target decision table first to determine a first candidate set of fault elements, then the fault diagnosis prediction model corresponding to each candidate fault element is obtained based on the first candidate set, the fault credibility corresponding to each candidate fault element is determined through the fault diagnosis prediction model, finally the fuzzy integral value corresponding to each candidate fault element is determined based on the fault credibility set, and the element of the power distribution network that has a fault is determined based on the fuzzy integral value, the target decision table is used for preliminary screening first, and then the fault diagnosis prediction model is intervened based on the preliminary screening result to determine the fault credibility, thereby avoiding the problems of large amount of calculation and slow fault positioning speed caused by the judgment of the fault diagnosis prediction model for each power element in the traditional fault detection, and finally the element of the power distribution network that has a fault is determined through the fuzzy integral fusion method based on the fault credibility of each candidate fault element, the accuracy of the determined power element that has a fault is ensured, and therefore the fault of the power distribution network is actively analyzed and judged accurately and quickly. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 A step flow chart of a power distribution network fault active analysis method is provided for the embodiment of the application.

[0022] Figure 2 A structural schematic diagram of a power distribution network fault active analysis terminal is provided for the embodiment of the application. DETAILED DESCRIPTION

[0023] To explain the technical content, the achieved purposes and effects of the application in detail, the following is explained in combination with the embodiments and the drawings.

[0024] Please refer to Figure 1 A power distribution network fault active analysis method, characterized in that, comprising the steps of:

[0025] S1, acquiring the topological structure information of the power distribution network, and constructing a corresponding target decision table according to the topological structure information, each record in the target decision table including a condition attribute and the corresponding result attribute;

[0026] S2, acquiring the fault feature of the fault position in the power distribution network, determining the matched condition attribute from the target decision table according to the fault feature, and determining the corresponding target result attribute according to the matched condition attribute;

[0027] S3, determining a first candidate set of fault elements according to the target result attribute, and acquiring the corresponding fault diagnosis prediction model of each candidate fault element in the first candidate set.

[0028] S4, determining a fault credibility corresponding to each candidate fault element according to the fault diagnosis prediction model, and forming a fault credibility set;

[0029] S5, determining a fuzzy integral value corresponding to each candidate fault element based on the fault credibility set, and determining the element of the power distribution network that has a fault according to the fuzzy integral value.

[0030] From the above description, the beneficial effects of the present application are that: according to the topology structure of the power distribution network, a corresponding target decision table is constructed, the target decision table contains each condition attribute and its corresponding result attribute, when the fault characteristics of the fault position in the power distribution network are obtained, the target decision table is matched according to the fault characteristics first to determine a first candidate set of fault elements, then the fault diagnosis prediction model corresponding to each candidate fault element is obtained based on the first candidate set, the fault credibility corresponding to each candidate fault element is determined through the fault diagnosis prediction model, finally the fuzzy integral value corresponding to each candidate fault element is determined based on the fault credibility set, and the element of the power distribution network that has a fault is determined based on the fuzzy integral value, the target decision table is used for preliminary screening, and the fault diagnosis prediction model is used for intervention based on the preliminary screening result to determine the fault credibility, thereby avoiding the problem of large amount of calculation and slow fault positioning speed caused by the judgment of the fault diagnosis prediction model for each power element in the traditional fault detection, and finally the element of the power distribution network that has a fault is determined through the fuzzy integral fusion method based on the fault credibility of each candidate fault element, the accuracy of the determined power element that has a fault is ensured, and thus the active research and judgment of the fault of the power distribution network is realized accurately and quickly.

[0031] Further, the constructing of the corresponding target decision table according to the topology structure information comprises:

[0032] constructing a corresponding initial decision table according to the topology structure information, reducing the initial decision table to obtain a minimal reduction table, and determining the minimal reduction table as the target decision table.

[0033] constructing a corresponding initial decision table according to the topology structure information, reducing the initial decision table to obtain a minimal reduction table, and determining the minimal reduction table as the target decision table.

[0034] From the above description, after the corresponding initial decision table is constructed according to the topology structure information, the initial decision table is reduced to obtain a minimal reduction table, and the minimal reduction table is determined as the target decision table, thereby avoiding the redundancy and complexity of the target decision table, and facilitating the rapidity and accuracy of subsequent matching of candidate fault elements.

[0035] Further, the S3 further comprises the steps of:

[0036] determining all the elements to be diagnosed of the power distribution network according to the result attribute of each record in the target decision table.

[0037] constructing a corresponding fault diagnosis prediction model for each element to be diagnosed;

[0038] training the fault diagnosis prediction model to obtain a trained fault diagnosis prediction model, forming a fault diagnosis prediction model library;

[0039] In S3, a corresponding fault diagnosis prediction model is obtained from the fault diagnosis prediction model library based on each candidate fault element in the first candidate set.

[0040] As can be seen from the above description, after the target decision table is determined, all elements to be diagnosed in the power distribution network are determined based on the result attribute in the target decision table, and then a corresponding fault diagnosis prediction model is constructed for each diagnostic element. After training, the fault diagnosis prediction model is added to the fault diagnosis prediction model library to facilitate subsequent direct calling after the first candidate fault element set is formed.

[0041] Further, S2 further comprises:

[0042] If the number of target result attributes is one, the element corresponding to the target result attribute is directly determined as the element that has failed in the power distribution network, otherwise, steps S3 to S5 are executed.

[0043] As can be seen from the above description, if there is only one candidate fault element located according to the target decision table, the candidate fault element is directly determined as the element that has failed in the power distribution network. Only when there is more than one candidate fault element determined, steps S3 to S5 are executed, further improving the efficiency of locating the fault element in the power distribution network.

[0044] Further, the determination of the element that has failed in the power distribution network based on the fuzzy integral value comprises:

[0045] A target candidate fault element with the maximum fuzzy integral value is determined, and the target candidate fault element is determined as the element that has failed in the power distribution network.

[0046] As can be seen from the above description, the candidate fault element with the maximum fuzzy integral value is determined as the element that has failed in the power distribution network, which can ensure the accuracy of the determined fault element.

[0047] Please refer to Figure 2 A power distribution network fault active research terminal, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0048] S1, acquire topology structure information of a power distribution network, and construct a corresponding target decision table according to the topology structure information, each record in the target decision table including a condition attribute and a corresponding result attribute;

[0049] S2, acquire a fault feature of a fault occurrence position in the power distribution network, determine a matched condition attribute from the target decision table according to the fault feature, and determine a corresponding target result attribute according to the matched condition attribute;

[0050] S3, determine a first candidate set of fault elements according to the target result attribute, and acquire a corresponding fault diagnosis prediction model based on each candidate fault element in the first candidate set;

[0051] S4, determine a fault credibility corresponding to each candidate fault element according to the fault diagnosis prediction model, and form a fault credibility set;

[0052] S5, determine a fuzzy integral value corresponding to each candidate fault element based on the fault credibility set, and determine the element of the power distribution network that has a fault according to the fuzzy integral value.

[0053] From the above description, the beneficial effects of the present application are that: according to the topology structure of the power distribution network, a corresponding target decision table is constructed, the target decision table contains each condition attribute and its corresponding result attribute, when the fault feature of the fault occurrence position in the power distribution network is acquired, the target decision table is matched according to the fault feature first, a first candidate set of fault elements is determined, then a fault diagnosis prediction model corresponding to each candidate fault element is acquired based on the first candidate set, the fault credibility corresponding to each candidate fault element is determined through the fault diagnosis prediction model, finally the fuzzy integral value corresponding to each candidate fault element is determined based on the fault credibility set, and the element of the power distribution network that has a fault is determined based on the fuzzy integral value, the target decision table is used for preliminary screening first, and then the fault diagnosis prediction model is intervened based on the preliminary screening result to determine the fault credibility, thereby avoiding the problem of large amount of calculation and slow fault positioning speed caused by the judgment of the fault diagnosis prediction model for each power element in the traditional fault detection, finally the element of the power distribution network that has a fault is determined through the fuzzy integral fusion method based on the fault credibility of each candidate fault element, the accuracy of the determined power element that has a fault is ensured, thereby realizing the active research and judgment of the fault of the power distribution network accurately and quickly.

[0054] Further, the construction of the corresponding target decision table according to the topology structure information includes:

[0055] An initial decision table is constructed according to the topology structure information, the initial decision table is reduced to obtain a minimum reduction table;

[0056] determining the minimum reduction table as the target decision table.

[0057] As can be seen from the above description, after the initial decision table corresponding to the topological structure information is constructed, the initial decision table is reduced to obtain the minimum reduction table, and the minimum reduction table is determined as the target decision table, thereby avoiding the redundancy and complexity of the target decision table and facilitating the rapidity and accuracy in subsequent matching of the candidate faulty components.

[0058] Further, the S3 further comprises the steps of:

[0059] determining all the components to be diagnosed in the power distribution network according to the result attribute of each record in the target decision table;

[0060] constructing a corresponding fault diagnosis prediction model for each component to be diagnosed;

[0061] training the fault diagnosis prediction model to obtain a trained fault diagnosis prediction model, thereby forming a fault diagnosis prediction model library;

[0062] In the S3, a corresponding fault diagnosis prediction model is obtained from the fault diagnosis prediction model library based on each candidate faulty component in the first candidate set.

[0063] As can be seen from the above description, after the target decision table is determined, all the components to be diagnosed in the power distribution network are determined based on the result attribute in the target decision table, and then a corresponding fault diagnosis prediction model is constructed for each diagnosed component. After training, the fault diagnosis prediction model is added to the fault diagnosis prediction model library to facilitate the subsequent direct calling after the first candidate faulty component set is formed.

[0064] Further, the S2 further comprises:

[0065] determining whether the number of target result attributes is one. If the number is one, the component corresponding to the target result attribute is directly determined as the faulty component in the power distribution network. Otherwise, steps S3 to S5 are executed.

[0066] As can be seen from the above description, if there is only one candidate faulty component located according to the target decision table, the candidate faulty component is directly determined as the faulty component in the power distribution network. Only when there is more than one candidate faulty component, steps S3 to S5 are started to be executed, thereby further improving the efficiency of locating the faulty component in the power distribution network.

[0067] Further, the determining of the faulty component in the power distribution network according to the fuzzy integral value comprises:

[0068] determining a target candidate faulty component with the maximum fuzzy integral value, and determining the target candidate faulty component as the faulty component in the power distribution network.

[0069] From the above description, it can be known that the candidate fault element with the maximum fuzzy integral value is determined as the fault element of the power distribution network, which can ensure the accuracy of the determined fault element.

[0070] The power distribution network fault active research method and the terminal described above can be applied to the research of faults in the power distribution network, which will be described below through specific embodiments:

[0071] Embodiment one

[0072] Please refer to Figure 1 A power distribution network fault active research method, comprising the steps of:

[0073] S1, obtaining the topology structure information of the power distribution network, and constructing a corresponding target decision table according to the topology structure information, each record in the target decision table including a condition attribute and its corresponding result attribute;

[0074] For example, the power distribution network can be divided into multiple power supply areas, and different power supply areas are interconnected by multiple switches, transformers, buses and lines, i.e. the topology structure information of the power distribution network is formed by the interconnection between multiple switches, transformers, buses and lines. At this time, the working state corresponding to each power supply area can be determined according to whether each switch, each transformer, each bus and each line is faulty, thereby forming a decision table taking the working state corresponding to each power supply area as the condition attribute and taking whether each electrical element of the power distribution network is faulty as the result attribute;

[0075] Among them, the construction of the corresponding target decision table according to the topology structure information includes:

[0076] According to the topology structure information, an initial decision table is constructed, and the initial decision table is reduced to obtain a minimum reduction table;

[0077] Among them, the existing immune algorithm or inheritance algorithm can be used to realize the reduction of the initial decision table;

[0078] The minimum reduction table is determined as the target decision table;

[0079] In an optional embodiment, when the topology structure of the power distribution network changes, the target decision table can be updated based on the changed topology structure information to realize dynamic updating of the target decision table to adapt to the change of the power distribution network;

[0080] S2, obtaining the fault characteristics of the fault position in the power distribution network, determining the matching condition attribute from the target decision table according to the fault characteristics, and determining the corresponding target result attribute according to the matching condition attribute;

[0081] For example, the determined target decision table is shown in Table 1:

[0082] Table 1

[0083]

[0084] In Table 1, 1 in the condition attribute represents that the corresponding power supply area is in a normal working state, and 0 represents that the corresponding power supply area is in an abnormal working state.

[0085] If the fault feature of the obtained fault occurrence position of the power distribution network is that the attribute value corresponding to the power supply area 2 is 0, the first record, the second record and the fourth record can be matched through Table 1, and thus it can be determined that the corresponding result attribute is the fault element: switch 1, transformer 1 and transformer 2.

[0086] The number of the target result attribute is determined, if the number is one, the element corresponding to the target result attribute is directly determined as the element of the power distribution network that has failed, otherwise, steps S3 to S5 are executed.

[0087] For example, in Table 1, if the fault feature of the obtained fault occurrence position of the power distribution network is that the attribute value corresponding to the power supply area 4 is 0, the fifth record in Table 1 is matched, and the corresponding result attribute is only one, that is, bus 1, and thus it can be directly determined that the fault element in the power distribution network is bus 1, and steps S3-S5 do not need to be executed; and if it is the above example, the number of result attributes is 3, and steps S3-S5 need to be executed.

[0088] S3, determining a first candidate set of fault elements according to the target result attribute, and obtaining a corresponding fault diagnosis prediction model based on each candidate fault element in the first candidate set.

[0089] For example, taking the attribute value corresponding to the power supply area 2 in the fault feature of the fault occurrence position in Table 1 as an example, the first candidate set can be obtained as {switch 1, transformer 1, transformer 2}, and at this time, the fault diagnosis prediction models corresponding to switch 1, transformer 1 and transformer 2 are obtained in turn.

[0090] The S3 further includes the following steps:

[0091] All the to-be-diagnosed elements of the power distribution network are determined according to the result attribute of each record in the target decision table.

[0092] A corresponding fault diagnosis prediction model is constructed for each to-be-diagnosed element.

[0093] The fault diagnosis prediction model is trained to obtain a trained fault diagnosis prediction model, and a fault diagnosis prediction model library is formed.

[0094] That is, a corresponding fault diagnosis prediction model can be constructed for each element in the power distribution network, and trained to obtain a trained fault diagnosis prediction model. Specifically, an RBF neural network model can be constructed. If it is a line, a line RBF neural network model is constructed. If it is a bus, a bus RBF neural network model is constructed. If it is a transformer, a transformer RBF neural network model is constructed.

[0095] The S3 obtains a corresponding fault diagnosis prediction model from the fault diagnosis prediction model library based on each candidate fault element in the first candidate set.

[0096] S4, according to the fault diagnosis prediction model, determine the fault credibility of each candidate fault element, and form a fault credibility set.

[0097] S5, based on the fault credibility set, determine the fuzzy integral value corresponding to each candidate fault element, and determine the element of the power distribution network that has failed according to the fuzzy integral value.

[0098] In a specific implementation, the Sugeno fuzzy integral or Choquet fuzzy integral commonly used in fuzzy integral technology can be used to determine the fuzzy integral value corresponding to each candidate fault element based on the fault credibility set.

[0099] The determination of the element of the power distribution network that has failed according to the fuzzy integral value comprises:

[0100] Determine the target candidate fault element with the maximum fuzzy integral value, and determine the target candidate fault element as the element of the power distribution network that has failed.

[0101] In another optional implementation, multiple fault elements can occur, so the candidate fault element with a fuzzy integral value greater than a preset threshold can be determined as the element of the power distribution network that has failed.

[0102] Embodiment two

[0103] Please refer to Figure 2 A power distribution network fault active research terminal, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement each step of the power distribution network fault active research method of embodiment one.

[0104] In summary, the application provides a kind of distribution network fault active research method and terminal, according to the topological structure of distribution network corresponding target decision table is constructed, target decision table is the minimum simple table, each condition attribute and its corresponding result attribute are contained in table, when the fault feature of the fault position in distribution network is obtained, first, according to the fault feature in target decision table is matched, the first candidate set of fault element is determined, then based on the first candidate set, the fault diagnosis prediction model corresponding to each candidate fault element is obtained, the fault credibility corresponding to each candidate fault element is determined through fault diagnosis prediction model, finally, based on the fault credibility set, the fuzzy integral value corresponding to each candidate fault element is determined, the element of distribution network fault is determined based on fuzzy integral value, first, through the minimum simple table is preliminarily screened, then, based on the preliminary screening result, the intervention of fault diagnosis prediction model is carried out to determine fault credibility, avoid the problem of large amount of calculation and slow fault positioning speed caused by the judgment of fault diagnosis prediction model of each power element in traditional fault detection, finally, based on the fault credibility of each candidate fault element, the final element of distribution network fault is determined by the method of fuzzy integral fusion, ensure the accuracy of the determined power element of fault, the decision-making judgment, the fault credibility prediction through fault diagnosis prediction model and fuzzy integral technology are combined, the accurate and rapid active research on the fault of distribution network is realized.

[0105] The above is only an embodiment of the present application, and does not limit the patent scope of the present application, any equivalent transformation or direct or indirect application in related technical field using the content of the present application specification and drawings is also included in the patent protection scope of the present application.

Claims

1. A power distribution network fault active research method, characterized in that, The method comprises the steps of: S1, obtaining topology structure information of a power distribution network, and constructing a corresponding target decision table according to the topology structure information, each record in the target decision table comprising a condition attribute and a corresponding result attribute; S2, obtaining a fault feature of a fault occurrence position in the power distribution network, determining a matched condition attribute from the target decision table according to the fault feature, and determining a corresponding target result attribute according to the matched condition attribute; S3, determining a first candidate set of fault elements according to the target result attribute, and obtaining a corresponding fault diagnosis prediction model based on each candidate fault element in the first candidate set; S4, determining a fault credibility corresponding to each candidate fault element according to the fault diagnosis prediction model, and forming a fault credibility set; S5, determining a fuzzy integral value corresponding to each candidate fault element based on the fault credibility set, and determining an element of the power distribution network that has a fault according to the fuzzy integral value; The step of constructing the corresponding target decision table according to the topology structure information comprises: constructing a corresponding initial decision table according to the topology structure information, reducing the initial decision table, and obtaining a minimum reduction table; determining the minimum reduction table as the target decision table; The step S3 further comprises the steps of: determining all to-be-diagnosed elements of the power distribution network according to the result attribute of each record in the target decision table; constructing a corresponding fault diagnosis prediction model for each to-be-diagnosed element; training the fault diagnosis prediction model to obtain a trained fault diagnosis prediction model, and forming a fault diagnosis prediction model library; In the step S3, the corresponding fault diagnosis prediction model of each candidate fault element in the first candidate set is obtained from the fault diagnosis prediction model library.

2. The method of claim 1, wherein, The step S2 further comprises: determining the number of target result attributes, if the number is one, directly determining the element corresponding to the target result attribute as the element of the power distribution network that has a fault, otherwise, executing steps S3 to S5.

3. The method of claim 1, wherein, The step of determining the element of the power distribution network that has a fault according to the fuzzy integral value comprises: determining a target candidate fault element with the maximum fuzzy integral value, and determining the target candidate fault element as the element of the power distribution network that has a fault.

4. An active fault research and judgment terminal for a power distribution network, comprising a memory, a processor, and a computer program stored on the memory and capable of running on the processor, characterized in that, The processor executes the computer program to implement the following steps: S1, obtaining topology structure information of a power distribution network, and constructing a corresponding target decision table according to the topology structure information, each record in the target decision table comprising a condition attribute and a corresponding result attribute; S2, obtaining a fault feature of a fault occurrence position in the power distribution network, determining a matched condition attribute from the target decision table according to the fault feature, and determining a corresponding target result attribute according to the matched condition attribute; S3, determining a first candidate set of fault elements according to the target result attribute, and obtaining a corresponding fault diagnosis prediction model based on each candidate fault element in the first candidate set; S4, determining a fault credibility corresponding to each candidate fault element according to the fault diagnosis prediction model, and forming a fault credibility set; S5, determining a fuzzy integral value corresponding to each candidate fault element based on the set of fault credibility, and determining the element of the power distribution network in which the fault occurs according to the fuzzy integral value; The step of constructing the corresponding target decision table according to the topological structure information comprises: constructing a corresponding initial decision table according to the topological structure information, reducing the initial decision table, and obtaining a minimum reduction table; determining the minimum reduction table as the target decision table; The step S3 further comprises the steps of: determining all the elements to be diagnosed of the power distribution network according to the result attribute of each record in the target decision table; constructing a corresponding fault diagnosis prediction model for each element to be diagnosed; training the fault diagnosis prediction model to obtain a trained fault diagnosis prediction model, and forming a fault diagnosis prediction model library; In the step S3, a corresponding fault diagnosis prediction model is obtained from the fault diagnosis prediction model library based on each candidate fault element in the first candidate set.

5. The power distribution network fault active research terminal according to claim 4, characterized in that, The step S2 further comprises the steps of: determining the number of the target result attributes, if the number is one, directly determining the element corresponding to the target result attribute as the element of the power distribution network in which the fault occurs, otherwise, executing the steps S3 to S5.

6. The power distribution network fault active research terminal according to claim 4, characterized in that, The step of determining the element of the power distribution network in which the fault occurs according to the fuzzy integral value comprises: determining a target candidate fault element with the maximum fuzzy integral value, and determining the target candidate fault element as the element of the power distribution network in which the fault occurs.

Images