A method and system for locating fault sections in a multi-source distribution network

Abstract

The present invention proposes a method and system for locating a fault section of a multi-source distribution network, the method comprising: establishing a linked list distribution network structure model; the model uses the distribution network feeder section as a node unit or the busbar as a node unit, and is connected according to the distribution network structure through a pointer, and the arrow of the pointer represents the direction of the flow; based on the model, a Bayesian network model is established based on multi-source fault alarm information, and a preliminary fault judgment is made through the Bayesian network model to determine the central section of the suspected fault area; according to the location of the fault, the boundary of the central section of the suspected fault area is determined to obtain the suspected fault area, and then the fault location is determined by analyzing the discreteness of the association weight between the alarm information and the suspected fault area. Based on this method, the present invention also proposes a multi-source distribution network fault section positioning system. The present invention uses the discreteness model of the association weight between the alarm information and the suspected fault section to determine the fault location, and the fault section is accurately located.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network fault location, and in particular relates to a method and system for locating a fault section of a multi-source distribution network. Background Art

[0002] With the advent of the smart grid era, the distribution network automation system has developed rapidly, and there are more and more distribution network fault location methods based on the distribution network automation system. The distribution network fault location method needs to adapt to the development of the distribution network automation system and meet the economic and social needs for power quality, which has always been a hot topic of research.

[0003] At present, there are mainly the following methods for locating distribution network faults: one is based on sound alarm information; the other is a distribution network fault location method based on artificial intelligence algorithms. However, for some areas, the distribution network automation system is not perfect, and there is no complete fault information collection system. These areas mainly use cheap fault indicators and distribution transformer alarm devices as the main fault alarm equipment, and use user power outage telephone complaint information to assist in fault location. Because it is impossible to collect complete electrical quantity fault information, there are few fault location methods suitable for this situation. The shortcomings of the existing scheme include: first, the fault calculation is complex, and there are many distribution network equipment, which brings huge pressure on the calculation amount, and the designed method is too idealized for actual application. Second, the fault judgment data is redundant. The amount of fault judgment data is huge, and the pressure on the storage and calculation of raw data, process data, and final data is huge. Third, the fault location data is idealized. In some areas, a more comprehensive monitoring system is configured, and the fault data is relatively limited, and it is impossible to locate this situation. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a method and system for locating a fault section in a multi-source distribution network, which can accurately locate the fault section and is applicable to complex and diverse distribution network systems.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for locating a fault section of a multi-source distribution network comprises the following steps:

[0007] Establishing a linked list distribution network structure model; the linked list distribution network structure model uses the distribution network feeder section as a node unit or the bus as a node unit, is connected according to the distribution network structure through a pointer, and the arrow of the pointer represents the direction of the flow;

[0008] Based on the linked list distribution network structure model, a Bayesian network model is established based on multi-source fault alarm information, and a preliminary fault judgment is made through the Bayesian network model to determine the central section of the suspected fault area;

[0009] According to the location of the fault, the boundary of the central section of the suspected fault area is determined to obtain the suspected fault area, and then the fault location is determined by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area.

[0010] Furthermore, the data domain of the node unit is used to store power distribution information of the feeder section or bus; the power distribution information includes status information of the section switch connected to the feeder section or bus, status information of the circuit breaker connected to the feeder section or bus, information of the distribution transformer belonging to the feeder section or bus, and load information of the corresponding user area.

[0011] Furthermore, the multi-source fault alarm information includes: first alarm information obtained by detecting the magnitude of the fault current through a fault indicator, second alarm information obtained by detecting whether the distribution transformer has lost power through a distribution transformer alarm device, and third alarm information obtained by a user's telephone complaint after a power outage.

[0012] Furthermore, the process of performing preliminary fault judgment through the Bayesian network model and determining the suspected fault area includes: judging whether the distribution transformer fault or the line fault occurs through the Bayesian network model; if the distribution transformer fault occurs, it can be directly located; if the line fault occurs, determining the central section of the suspected fault area to form a suspected fault area.

[0013] Furthermore, the process of determining the boundary of the central section of the suspected fault area according to the location of the fault to obtain the boundary of the suspected fault area includes: defining the node unit at the end of the arrow as the downstream node of the starting node unit of the arrow in the linked list distribution network structure model, and the starting node unit as the upstream node of the downstream node unit; the upstream node and downstream nodes of the central section of the suspected fault area are the boundary of the suspected fault area.

[0014] Furthermore, after determining the boundary of the suspected fault area, it also includes establishing a mapping relationship between the alarm device and the fault location corresponding to the alarm device, and defining it as an association relationship.

[0015] Furthermore, the process of determining the fault location by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area includes:

[0016] Calculate the alarm information association information value of each possible fault feeder section in the suspected fault area, and determine the association degree ratio matrix according to the alarm information association information value; calculate the maximum eigenvalue and corresponding eigenvector of the association degree ratio matrix; the eigenvector is the association weight between the feeder section and the corresponding alarm information;

[0017] The ratio of the standard deviation coefficient of the associated weight to the average value of the associated weight is used as the discrete coefficient of the associated weight of the feeder section. Through Hermite interpolation, the discrete coefficient of the associated weight of the feeder section is used as a known function value to form a discrete coefficient variation diagram of the suspected fault area. The minimum point of the discrete coefficient variation diagram of the suspected fault area is the fault feeder position.

[0018] The present invention also proposes a multi-source distribution network fault section positioning system, including a model building module, a preliminary judgment module and a fault determination module;

[0019] The model building module is used to build a linked list distribution network structure model; the linked list distribution network structure model uses the distribution network feeder section as a node unit or the bus as a node unit, is connected according to the distribution network structure through a pointer, and the arrow of the pointer represents the direction of the flow;

[0020] The preliminary judgment module is used to establish a Bayesian network model based on the linked list distribution network structure model and multi-source fault alarm information, and to make a preliminary fault judgment through the Bayesian network model to determine the central section of the suspected fault area;

[0021] The fault determination module is used to determine the boundary of the central section of the suspected fault area according to the fault occurrence location to obtain the suspected fault area, and then determine the fault location by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area.

[0022] Furthermore, the execution process of the preliminary judgment module includes: based on the linked list distribution network structure model, a Bayesian network model is established based on multi-source fault alarm information, and through the Bayesian network model, it is determined whether the distribution transformer fault or the line fault occurs; if the distribution transformer fault occurs, it can be directly located; if the line fault occurs, the central section of the suspected fault area is determined to form a suspected fault area.

[0023] Further, the process of determining the execution of the fault module includes:

[0024] In the linked list distribution network structure model, the node unit at the end of the arrow is defined as the downstream node of the arrow start node unit, and the start node unit is defined as the upstream node of the downstream node unit; the upstream node and downstream node of the central section of the suspected fault area are the boundaries of the suspected fault area.

[0025] Establish a mapping relationship between the alarm device and the fault location corresponding to the alarm device, and define it as an association relationship;

[0026] Calculate the alarm information association information value of each possible faulty feeder section in the suspected fault area, and determine the association degree ratio matrix according to the alarm information association information value; calculate the maximum eigenvalue and the corresponding eigenvector of the association degree ratio matrix; the eigenvector is the association weight between the feeder section and the corresponding alarm information; use the ratio of the standard deviation coefficient of the association weight to the average value of the association weight as the association weight discrete coefficient of the feeder section; through Hermite interpolation, use the association weight discrete coefficient of the feeder section as the known function value to form a discrete coefficient change diagram of the suspected fault area; the minimum point of the discrete coefficient change diagram of the suspected fault area is the fault feeder position.

[0027] The effects provided in the content of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:

[0028] The present invention proposes a method and system for locating a fault section of a multi-source distribution network, the method comprising the following steps: establishing a linked list distribution network structure model; the linked list distribution network structure model uses the distribution network feeder section as a node unit or the busbar as a node unit, and is connected according to the distribution network structure through a pointer, and the arrow of the pointer represents the direction of the current; based on the linked list distribution network structure model, a Bayesian network model is established based on multi-source fault alarm information, and a preliminary fault judgment is performed through the Bayesian network model to determine the central section of the suspected fault area; according to the location of the fault, the boundary of the central section of the suspected fault area is determined to obtain the suspected fault area, and then the fault location is determined by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area. Based on a method for locating a fault section of a multi-source distribution network, the present invention also proposes a system for locating a fault section of a multi-source distribution network. The present invention adopts a distribution network linked list modeling method, the original data is simple, and the method can be applied to areas where the distribution network system is not comprehensive. The Bayesian network model is established based on multi-source fault alarm information, the calculation pressure is small, and the calculation and storage pressure of the distribution network system is relieved. The discrete model of the correlation weight between the alarm information and the suspected fault section determines the fault location, and the fault section is accurately located, which is suitable for complex and diverse distribution network systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] like Figure 1 This is a flow chart of a method for locating a fault section in a multi-source distribution network according to Embodiment 1 of the present invention;

[0030] like Figure 2 This is a schematic diagram of the Bayesian model of Example 1 of the present invention;

[0031] like Figure 3 This is a schematic diagram of a multi-source distribution network fault section locating system according to Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0032] In order to clearly illustrate the technical features of the present solution, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present invention omits the description of known components and processing techniques and processes to avoid unnecessary limitations on the present invention.

[0033] Example 1

[0034] Embodiment 1 of the present invention proposes a method for locating a fault section of a multi-source distribution network, which is used to solve the problems of complex fault calculation and inaccurate fault location in the prior art.

[0035] like Figure 1 This is a flow chart of a method for locating a fault section in a multi-source distribution network according to Embodiment 1 of the present invention;

[0036] In step S100, the process starts;

[0037] In step S110, a distribution network structure diagram is formed;

[0038] In step S120, a linked list distribution network structure model is established. Currently, most distribution networks are closed-loop structures with open-loop operation, which have many line components and complex branch lines, making it difficult to locate distribution network faults. The linked list structure is simple and can well describe the component information in the distribution network. At the same time, the pointer of the linked list can show the connection relationship between the components of the distribution network, which is suitable for the complex structure of the distribution network.

[0039] The linked list distribution network structure model uses the distribution network feeder section as the node unit or the busbar as the node unit, which is connected according to the distribution network structure through pointers, and the arrow of the pointer represents the direction of the flow;

[0040] The data domain of the node unit is used to store the power distribution information of the feeder section or bus; the power distribution information includes the status information of the sectionalized switch connected to the feeder section or bus, the status information of the circuit breaker connected to the feeder section or bus, the information of the distribution transformer to which the feeder section or bus belongs, and the load information of the corresponding user area. If there are n distribution network feeder sections in the distribution network, a linked list model with n node units will be formed.

[0041] In step S130, based on the linked list distribution network structure model, a Bayesian network model is established based on multi-source fault alarm information.

[0042] When a fault occurs somewhere in the distribution network, the fault alarm device will generate an alarm message, which generally includes:

[0043] The fault indicator sends an alarm by detecting the magnitude of the fault current. When a fault occurs downstream of the fault indicator of a section switch, the fault indicator of the section switch will detect the fault current and send an alarm message; while the fault indicator downstream of the fault point will not send an alarm message because it does not detect the occurrence of the fault current.

[0044] The distribution transformer alarm device alarms by detecting whether the distribution transformer loses power. After the fault occurs, the faulty feeder section will be isolated through the switching operation, and the distribution transformer connected to the faulty feeder section will lose power, so that the distribution transformer alarm device at the distribution transformer will alarm. The distribution transformer alarm information contains the status of the fuse switch on the high-voltage side of the distribution transformer. According to the status of the fuse switch, it can be distinguished whether it is a line fault or a distribution transformer fault.

[0045] User telephone complaint information comes from the power outage telephone complaints of users after the power outage. Since user telephone complaint information has strong randomness and uncertainty, user telephone complaint information is only used as auxiliary information for the other two fault alarm information. Through user telephone complaint information, the power outage area can be determined first, and then the fault point can be determined through the structural logical relationship of the distribution network.

[0046] The suspected fault area is divided into the suspected fault center section and the suspected fault area boundary. A preliminary fault location is performed by establishing a Bayesian network model to determine whether the distribution transformer or line fault occurs. If the distribution transformer fails, it can be directly located; if the line fails, the suspected fault area center section can be determined at the same time to form a suspected fault area.

[0047] Bayesian networks require that the events that constitute a complete event group are independent of each other. The fault indicator alarm information and the distribution transformer alarm information come from different fault alarm devices at the line section switch and the distribution transformer, respectively, and are independent of each other; at the same time, the user telephone complaint information is independent of the above two types of alarm information, which meets the conditions for the application of Bayesian networks. Figure 2 Schematic diagram of the Bayesian model of Embodiment 1 of the present invention, wherein A1 represents a distribution network feeder section fault, A2 represents a distribution transformer body fault, B1 represents a node unit fault indicator alarm information, B2 represents a node unit distribution transformer alarm information, and B3 represents a node unit user telephone complaint information.

[0048] In step S140, a preliminary fault judgment is performed using the Bayesian network model to determine the central section of the suspected fault area; if it is a line fault, step S150 is executed, otherwise step S180 is executed.

[0049] In step S150, according to the location of the fault, the boundary of the central section of the suspected fault area is determined to obtain the suspected fault area.

[0050] Distribution line fault location can be divided into three parts. The first part is to form a suspected fault area based on the linked list distribution network model and preliminary fault judgment; the second part is the correlation relationship characterization; and the third part is to form a discrete coefficient change diagram of the fault area and determine the fault line section.

[0051] The Bayesian network model method can be used to determine the center of the suspected fault area while eliminating the distribution transformer fault. Due to the possibility of false alarm and missing fault alarm information, it is necessary to define the boundary of the suspected fault area to form a suspected fault area. In the distribution network model with a linked list structure, the node unit at the end of the arrow is defined as the downstream node of the arrow start node unit, and the start node unit is defined as the upstream node of the downstream node unit. The upstream and downstream nodes of the central section of the suspected fault area are the boundaries of the suspected fault area.

[0052] In step S160, the fault location is determined by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area. Specifically, the following steps are performed:

[0053] In the linked list distribution network structure model, the node unit at the end of the arrow is defined as the downstream node of the arrow start node unit, and the start node unit is defined as the upstream node of the downstream node unit; the upstream node and downstream node of the central section of the suspected fault area are the boundaries of the suspected fault area.

[0054] Establish a mapping relationship between the alarm device and the fault location corresponding to the alarm device, and define it as an association relationship;

[0055] Calculate the alarm information association information value of each possible faulty feeder section in the suspected fault area, and determine the association degree ratio matrix according to the alarm information association information value; calculate the maximum eigenvalue and the corresponding eigenvector of the association degree ratio matrix; the eigenvector is the association weight between the feeder section and the corresponding alarm information; use the ratio of the standard deviation coefficient of the association weight to the average value of the association weight as the association weight discrete coefficient of the feeder section; through Hermite interpolation, use the association weight discrete coefficient of the feeder section as the known function value to form a discrete coefficient change diagram of the suspected fault area; the minimum point of the discrete coefficient change diagram of the suspected fault area is the fault feeder position.

[0056] Each element g in the suspected fault region G i Indicates the possible faulty feeder section. i , the alarm information set W can be divided into the associated alarm information set W g and non-associated alarm information sets At the same time, the alarm information set W is associated g and non-associated alarm information sets It is divided into three types of alarm information, namely, the alarm information of the associated fault indicator W g1 , associated distribution transformer alarm information W g2 , associated user telephone complaint information W g3 ; Non-associated fault indicator alarm information Non-associated distribution transformer alarm information Non-associated user telephone complaint information

[0057] For each feeder section g i , randomly select a pair of alarm information w from the alarm information set W i and w j , for the alarm information w i and w j , let f(w i / w j ) indicates alarm information i Relative to the alarm information w j The information value of "correlation comparison" is f(w i / w j ) and f(w j / w i ) Information values ​​are shown in Table 1 below.

[0058]

[0059] make

[0060] Then the feeder section g i The correlation ratio matrix is

[0061]

[0062] where b ii =1,b ij b ji =1. The highest order of this correlation ratio matrix is ​​a 6-order matrix, and the matrix order varies with the alarm information. The correlation ratio matrix B is obtained by the characteristic equation: i The maximum eigenvalue λ max and the corresponding eigenvectors.

[0063] |B-lE|X=0

[0064] The eigenvector is the feeder section g i The associated weight with the corresponding alarm information is normalized when the eigenvector has a negative value. iThe correlation between the alarm information is quantified.

[0065] The dispersion coefficient describes the degree of dispersion on the unit mean and is often used to compare the dispersion of two population means with different dispersion. The dispersion coefficient of the associated weight is used to characterize different feeder sections g i The correlation between the alarm information and the faulty feeder section is high, and the dispersion coefficient is small, while the non-faulty feeder section is the opposite. This paper uses the ratio of the standard deviation coefficient of the correlation weight to its mean value as the feeder section g i The discrete coefficient of the associated weight is calculated as follows:

[0066]

[0067] LS i is the feeder section g i The coefficient of dispersion of the associated weight, S the standard deviation of the associated weight, The average of the associated weights.

[0068] By Hermite interpolation, the feeder section g i The discrete coefficient of the suspected fault area is used as the known function value to form a discrete coefficient change diagram of the suspected fault area. The discrete coefficient change diagram of the suspected fault area shows the feeder section g i If the alarm information has a high correlation with the faulty feeder section, the discrete coefficient is small, otherwise it is large. Therefore, in the discrete coefficient change diagram formed, the minimum value point is the faulty feeder section.

[0069] In step S170, the fault line segment area is determined.

[0070] In step S180 , a faulty line section is determined.

[0071] In step S190, the process ends.

[0072] A method for locating a fault section of a multi-source distribution network proposed in Example 1 of the present invention adopts a distribution network linked list modeling method, and the original data is simple. The method can be applied to areas where the distribution network system is not comprehensive. A Bayesian network model is established based on multi-source fault alarm information, which has a small amount of calculation pressure and alleviates the calculation and storage pressure of the distribution network system. The discrete model of the association weight between the alarm information and the suspected fault section determines the fault location, and the fault section is accurately located, which is suitable for complex and diverse distribution network systems.

[0073] Example 2

[0074] Based on a multi-source distribution network fault section positioning method proposed in Example 1 of the present invention, Example 2 of the present invention further proposes a multi-source distribution network fault section positioning system, such as Figure 3This is a schematic diagram of a multi-source distribution network fault section positioning system according to embodiment 2 of the present invention, the system comprising: a model building module, a preliminary judgment module and a fault determination module;

[0075] The model building module is used to build a linked list distribution network structure model; the linked list distribution network structure model uses the distribution network feeder section as a node unit or the bus as a node unit, is connected according to the distribution network structure through a pointer, and the arrow of the pointer represents the direction of the flow;

[0076] The preliminary judgment module is used to establish a Bayesian network model based on the linked list distribution network structure model and multi-source fault alarm information, and to make a preliminary fault judgment through the Bayesian network model to determine the central section of the suspected fault area;

[0077] The fault determination module is used to determine the boundary of the central section of the suspected fault area according to the location of the fault, and then determine the fault location by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area.

[0078] In the model building module, the linked list distribution network structure model uses the distribution network feeder section as the node unit or the busbar as the node unit, which is connected according to the distribution network structure through pointers, and the arrow of the pointer represents the direction of the flow;

[0079] The data domain of the node unit is used to store the power distribution information of the feeder section or bus; the power distribution information includes the status information of the sectionalized switch connected to the feeder section or bus, the status information of the circuit breaker connected to the feeder section or bus, the information of the distribution transformer to which the feeder section or bus belongs, and the load information of the corresponding user area. If there are n distribution network feeder sections in the distribution network, a linked list model with n node units will be formed.

[0080] The execution process of the preliminary judgment module includes: based on the linked list distribution network structure model, a Bayesian network model is established based on multi-source fault alarm information, and through the Bayesian network model, it is determined whether the distribution transformer fault or the line fault occurs; if the distribution transformer fault occurs, it can be directly located; if the line fault occurs, the central section of the suspected fault area is determined to form a suspected fault area.

[0081] The process of determining the faulty module execution includes:

[0082] In the linked list distribution network structure model, the node unit at the end of the arrow is defined as the downstream node of the arrow start node unit, and the start node unit is defined as the upstream node of the downstream node unit; the upstream node and downstream node of the central section of the suspected fault area are the boundaries of the suspected fault area.

[0083] Establish a mapping relationship between the alarm device and the fault location corresponding to the alarm device, and define it as an association relationship;

[0084] Calculate the alarm information association information value of each possible faulty feeder section in the suspected fault area, and determine the association degree ratio matrix according to the alarm information association information value; calculate the maximum eigenvalue and the corresponding eigenvector of the association degree ratio matrix; the eigenvector is the association weight between the feeder section and the corresponding alarm information; use the ratio of the standard deviation coefficient of the association weight to the average value of the association weight as the association weight discrete coefficient of the feeder section; through Hermite interpolation, use the association weight discrete coefficient of the feeder section as the known function value to form a discrete coefficient change diagram of the suspected fault area; the minimum point of the discrete coefficient change diagram of the suspected fault area is the fault feeder position.

[0085] A multi-source distribution network fault section positioning system proposed in Example 2 of the present invention adopts a distribution network linked list modeling method, the original data is simple, and the method can be applied to areas where the distribution network system is not comprehensive. A Bayesian network model is established based on multi-source fault alarm information, the calculation pressure is small, and the calculation and storage pressure of the distribution network system is alleviated. The alarm information and the discrete model of the association weight between the suspected fault section determine the fault location, and the fault section is accurately located, which is suitable for complex and diverse distribution network systems.

[0086] The description of the relevant parts of a multi-source distribution network fault section locating system provided in Example 2 of the present application can be found in the detailed description of the corresponding parts of a multi-source distribution network fault section locating method provided in Example 1 of the present application, and will not be repeated here.

[0087] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment that includes a series of elements are inherent to the elements. In the absence of more restrictions, the elements limited by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment that includes the elements. In addition, the above-mentioned technical solution provided in the embodiment of the present application is consistent with the corresponding technical solution in the prior art in principle, and the part is not described in detail, so as not to repeat too much.

[0088] Although the above describes the specific implementation of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. For those skilled in the art, other different forms of modifications or deformations can be made on the basis of the above description. It is not necessary and impossible to list all the implementation methods here. On the basis of the technical solution of the present invention, various modifications or deformations that can be made by those skilled in the art without creative work are still within the scope of protection of the present invention.

Claims

1. A method for locating a fault section in a multi-source distribution network, characterized in that: The following steps are involved: Establishing a linked list distribution network structure model; the linked list distribution network structure model uses the distribution network feeder section as a node unit or the bus as a node unit, is connected according to the distribution network structure through a pointer, and the arrow of the pointer represents the direction of the flow; Based on the linked list distribution network structure model, a Bayesian network model is established based on multi-source fault alarm information, and a preliminary fault judgment is made through the Bayesian network model to determine the central section of the suspected fault area; According to the location of the fault, the boundary of the central section of the suspected fault area is determined to obtain the suspected fault area, and then the fault location is determined by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area.

2. A method for locating a fault section of a multi-source distribution network according to claim 1, characterized in that: The data domain of the node unit is used to store the power distribution information of the feeder section or bus; the power distribution information includes the status information of the section switch connected to the feeder section or bus, the status information of the circuit breaker connected to the feeder section or bus, the distribution transformer information of the feeder section or bus and the load information of the corresponding user area.

3. A method for locating a fault section of a multi-source distribution network according to claim 1, characterized in that: The multi-source fault alarm information includes: first alarm information obtained by detecting the magnitude of the fault current through a fault indicator, second alarm information obtained by detecting whether the distribution transformer has lost power through a distribution transformer alarm device, and third alarm information obtained by a user's telephone complaint after a power outage.

4. A method for locating a fault section of a multi-source distribution network according to claim 1, characterized in that: The process of performing preliminary fault judgment through the Bayesian network model and determining the suspected fault area includes: judging whether the distribution transformer fault or the line fault occurs through the Bayesian network model; if the distribution transformer fault occurs, it can be directly located; if the line fault occurs, determining the central section of the suspected fault area to form a suspected fault area.

5. A method for locating a fault section of a multi-source distribution network according to claim 1, characterized in that: The process of determining the boundary of the central section of the suspected fault area according to the location of the fault to obtain the boundary of the suspected fault area includes: defining the node unit at the end of the arrow as the downstream node of the starting node unit of the arrow in the linked list distribution network structure model, and the starting node unit as the upstream node of the downstream node unit; the upstream node and downstream node of the central section of the suspected fault area are the boundary of the suspected fault area.

6. A method for locating a fault section of a multi-source distribution network according to claim 5, characterized in that: After determining the boundary of the suspected fault area, it also includes establishing a mapping relationship between the alarm device and the fault location corresponding to the alarm device, and defining it as an association relationship.

7. A method for locating a fault section of a multi-source distribution network according to claim 6, characterized in that: The process of determining the fault location by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area includes: Calculate the alarm information association information value of each possible fault feeder section in the suspected fault area, and determine the association degree ratio matrix according to the alarm information association information value; calculate the maximum eigenvalue and corresponding eigenvector of the association degree ratio matrix; the eigenvector is the association weight between the feeder section and the corresponding alarm information; The ratio of the standard deviation coefficient of the associated weight to the average value of the associated weight is used as the discrete coefficient of the associated weight of the feeder section. Through Hermite interpolation, the discrete coefficient of the associated weight of the feeder section is used as a known function value to form a discrete coefficient variation diagram of the suspected fault area. The minimum point of the discrete coefficient variation diagram of the suspected fault area is the fault feeder position.

8. A multi-source distribution network fault section location system, characterized in that: It includes model building module, preliminary judgment module and fault determination module; The model building module is used to build a linked list distribution network structure model; the linked list distribution network structure model uses the distribution network feeder section as a node unit or the bus as a node unit, is connected according to the distribution network structure through a pointer, and the arrow of the pointer represents the direction of the flow; The preliminary judgment module is used to establish a Bayesian network model based on the linked list distribution network structure model and multi-source fault alarm information, and to make a preliminary fault judgment through the Bayesian network model to determine the central section of the suspected fault area; The fault determination module is used to determine the boundary of the central section of the suspected fault area according to the fault occurrence location to obtain the suspected fault area, and then determine the fault location by analyzing the discreteness of the associated weights of the alarm information and the suspected fault area.

9. A multi-source distribution network fault section location system according to claim 8, characterized in that: The execution process of the preliminary judgment module includes: based on the linked list distribution network structure model, a Bayesian network model is established based on multi-source fault alarm information, and through the Bayesian network model, it is determined whether the distribution transformer fault or the line fault occurs; if the distribution transformer fault occurs, it can be directly located; if the line fault occurs, the central section of the suspected fault area is determined to form a suspected fault area.

10. A multi-source distribution network fault section location system according to claim 8, characterized in that: The process of determining the execution of the fault module includes: In the linked list distribution network structure model, the node unit at the end of the arrow is defined as the downstream node of the node unit at the start of the arrow, and the start node unit is defined as the upstream node of the downstream node unit; the upstream node and the downstream node of the central section of the suspected fault area are the boundaries of the suspected fault area; Establish a mapping relationship between the alarm device and the fault location corresponding to the alarm device, and define it as an association relationship; Calculate the alarm information association information value of each possible faulty feeder section in the suspected fault area, and determine the association degree ratio matrix according to the alarm information association information value; calculate the maximum eigenvalue and the corresponding eigenvector of the association degree ratio matrix; the eigenvector is the association weight between the feeder section and the corresponding alarm information; use the ratio of the standard deviation coefficient of the association weight to the average value of the association weight as the association weight discrete coefficient of the feeder section; through Hermite interpolation, use the association weight discrete coefficient of the feeder section as the known function value to form a discrete coefficient change diagram of the suspected fault area; the minimum point of the discrete coefficient change diagram of the suspected fault area is the fault feeder position.

Images