Power distribution network fault processing training method and device based on decision-making auxiliary model

Abstract

The invention discloses a power distribution network fault processing training method and device based on a decision-making auxiliary model. The method comprises the following steps: constructing a weight primitive library; obtaining a training sample set, wherein the training sample set comprises a plurality of power distribution network fault topology models; determining a plurality of simulation response plans, a plurality of corresponding standard weight prime combination schemes and standard tendency features based on the training sample set; training a power distribution network fault recovery decision auxiliary model based on the training sample set to obtain a plurality of simulation response plans, and a plurality of corresponding simulation weight prime combination schemes and simulation tendency features; and updating a first decision-making mechanism and a second decision-making mechanism of the power distribution network fault recovery decision-making auxiliary model. According to the invention, more reliable and accurate auxiliary information can be provided for a manager when a power grid has a fault in a later period. And meanwhile, multiple sets of personalized power supply recovery schemes are provided, so that the intelligent degree of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network, and in particular to a distribution network fault processing training method and device based on a decision support model. Background Art

[0002] As more and more electrical equipment is used in social life, society's requirements for grid stability are getting higher and higher. Accordingly, grid operators need to optimize the grid to avoid grid failures as much as possible, and repair grid failures in the shortest possible time when they occur. When a failure occurs, quickly and to the greatest extent possible to restore power supply, shorten or even eliminate the user's perception of power outage time, and improve the reliability of the distribution network have become one of the important tasks of grid construction and maintenance.

[0003] Traditional fault handling methods require technicians to analyze and make decisions based on the distribution pattern of the power grid and the grid fault warnings, which requires a high level of experience from the technicians and also prolongs the repair time of the grid fault. The automatic fault handling methods developed based on this usually adopt a general mode, cannot adaptively recommend fault handling solutions based on the affected equipment type, and are difficult to provide multiple sets of personalized power supply restoration solutions, and lack intelligence. The training method of the decision-making support model that overcomes the above defects has become an urgent problem to be solved. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention proposes a distribution network fault processing training method based on a decision support model, comprising:

[0005] Constructing a weight element library, wherein the weight element library includes a plurality of weight element types;

[0006] Assign corresponding tendency labels to each weight factor type;

[0007] Acquire a training sample set, wherein the training sample set includes a plurality of distribution network fault topology models, each distribution network fault topology model includes a full-circuit topology element, simulation parameters of each full-circuit topology element, and a simulated predetermined transfer line;

[0008] Determining a plurality of simulated response plans based on the training sample set, and a plurality of standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics;

[0009] Based on the training sample set, the distribution network fault recovery decision-making auxiliary model is trained to obtain a plurality of simulated response plans, a plurality of simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics;

[0010] Based on the tendency label, multiple standard weight element combination schemes and standard tendency characteristics, and corresponding simulated weight element combination schemes and simulated tendency characteristics, updating the first decision mechanism of the distribution network fault recovery decision auxiliary model;

[0011] Based on the distribution network fault topology model, multiple simulated response plans and corresponding standard response plans, the second decision mechanism of the distribution network fault recovery decision auxiliary model is updated.

[0012] In one embodiment, the above-mentioned training of the distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain multiple simulated response plans, and multiple simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics, include:

[0013] Based on the distribution network fault topology model, a simulated directed acyclic graph topology diagram is generated through a distribution network fault recovery decision-making auxiliary model, wherein the simulated directed acyclic graph topology diagram includes a simulated predetermined transfer line;

[0014] Based on the simulation parameters of all the topological elements of the entire road in the simulated directed acyclic graph topology diagram, the simulated fault equipment, simulated fault range and simulated fault accident type are determined.

[0015] In one embodiment, after determining the simulated fault device, simulated fault range and simulated fault accident type based on the simulation parameters of each full-road topology element in the simulated directed acyclic graph topology graph, the method includes:

[0016] Based on the device type of the affected device within the simulated fault range and the first decision mechanism, a corresponding weight element type is selected from the weight element library;

[0017] Based on the priority and quantity of the affected devices within the fault range, as well as the technical parameters and simulation parameters of the affected devices, a weight assignment operation is performed for the corresponding weight element type according to the first decision mechanism.

[0018] In one embodiment, after performing a weight assignment operation for the corresponding weight element type according to the first decision mechanism based on the priority and quantity of the affected devices within the fault range, as well as the technical parameters and real-time parameters of the affected devices, the above-mentioned method includes:

[0019] Determine multiple simulated response plans based on simulated fault equipment, simulated fault scope, simulated fault accident type, simulated scheduled transfer line and second decision-making mechanism;

[0020] Based on a plurality of corresponding weight element types, weight values ​​and a first decision mechanism, a plurality of simulation weight element combination schemes corresponding to each simulation response plan are generated.

[0021] In one embodiment, after generating multiple simulation weight element combination schemes corresponding to each simulation response plan based on multiple corresponding weight element types, weight values ​​and the first decision mechanism, the method includes:

[0022] Based on the corresponding simulation weight element combination schemes, the weight values ​​of each weight element type in the combination schemes, the tendency labels and the first decision-making mechanism, a simulation tendency feature is generated.

[0023] In one embodiment, the weight element library includes:

[0024] Whether it is a remotely controlled point for dispatching, the risk of voltage loss of the power outage equipment elements, the maximum current of the power outage equipment distribution network line in the month, the maximum current of the opposite ring network line in the month, the minimum value of the current carrying capacity of the opposite ring network line compared with the minimum current carrying capacity of the ring network loop, the number of users affected by the power outage equipment line, the number of users of the side distribution network line, the bus voltage difference where the line is located, the load rate of the main transformer and main network line and other main network equipment belonging to the opposite side line of the power transfer, the number of users affected by the power outage equipment bus voltage loss, the upper power supply can be traced back to the 10(20)kV ring network line that supplies power to the same 220kV substation, whether the line has completed phase sequence phase verification and trial closing test, whether it contains important sensitive users, the capacity-load ratio of transformers on both sides of the line, the secondary current of CT ratio on both sides of the line, the length of the power outage equipment distribution network line, the length of the opposite ring network line and the lines involved in the complaint work order.

[0025] The present invention also provides a distribution network fault processing training device based on a decision support model, comprising:

[0026] A construction module, used to construct a weight element library, wherein the weight element library includes a plurality of weight element types;

[0027] A matching module is used to assign corresponding tendency labels to each weight factor type;

[0028] An acquisition module is used to acquire a training sample set, wherein the training sample set includes a plurality of distribution network fault topology models, each of which includes a full-circuit topology element, simulation parameters of each full-circuit topology element, and a simulated predetermined transfer line;

[0029] A determination module, configured to determine a plurality of simulated response plans based on the training sample set, and a plurality of standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics;

[0030] A training module, used for training the distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain a plurality of simulated response plans, a plurality of simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics;

[0031] A first updating module, used to update a first decision mechanism of a distribution network fault recovery decision auxiliary model based on the tendency label, a plurality of standard weight element combination schemes and standard tendency characteristics, and corresponding simulation weight element combination schemes and simulation tendency characteristics;

[0032] The second updating module is used to update the second decision mechanism of the distribution network fault recovery decision auxiliary model based on the distribution network fault topology model, multiple simulation response plans and corresponding standard response plans.

[0033] The present invention also provides a computer device, a distribution network fault processing training system based on a decision-supporting model, the distribution network fault processing training system based on a decision-supporting model comprises: a distribution network fault processing device of a decision-supporting model;

[0034] The distribution network fault processing device of the decision support model is determined by a distribution network fault processing training device method based on a decision support model shown in any of the above items.

[0035] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned distribution network fault handling training method based on the decision support model when executing the computer program.

[0036] The present invention also provides a computer storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned distribution network fault processing training method based on the decision-making auxiliary model is implemented.

[0037] The embodiment of the present invention constructs a weight element library, which includes multiple weight element types; assigns corresponding tendency labels to each weight element type; obtains a training sample set, which includes multiple distribution network fault topology models, each distribution network fault topology model includes a full-line topology element, simulation parameters of each full-line topology element and a simulated scheduled transfer line; determines multiple simulated response plans based on the training sample set, and multiple standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics; trains a distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain multiple simulated response plans, and multiple simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics; based on the tendency labels, multiple standard weight element combination schemes and standard tendency characteristics, and corresponding simulated weight element combination schemes and simulated tendency characteristics, updates the first decision mechanism of the distribution network fault recovery decision-making auxiliary model; based on the distribution network fault topology model, multiple simulated response plans and corresponding standard response plans, updates the second decision mechanism of the distribution network fault recovery decision-making auxiliary model. The embodiment of the present invention can provide a first decision-making mechanism and a second decision-making mechanism that are continuously updated and improved through training, so that the determination of the weight factor combination scheme is more accurate and efficient, the simulated tendency characteristics can be as close to the standard tendency characteristics as possible, and the simulated response plan can be as close to the standard response plan as possible, providing managers with more reliable and accurate auxiliary information in the event of a later power grid failure. At the same time, the embodiment of the present invention can improve the intelligence of the system by providing multiple sets of personalized power supply restoration schemes. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 It is a flow chart of a distribution network fault processing method based on a decision support model according to a first embodiment of the present invention;

[0040] Figure 2 is a detailed flow chart of S11 of the first embodiment of the present invention;

[0041] Figure 3 is another detailed flow chart of S11 of the first embodiment of the present invention, which shows the distribution network fault processing training method based on the decision support model according to the embodiment of the present invention;

[0042] Figure 4is a detailed flow chart of S118 of the first embodiment of the present invention;

[0043] Figure 5 is a detailed flow chart of S12 of the first embodiment of the present invention;

[0044] Figure 6 It is an example diagram of constructing a directed acyclic topology graph according to a terminal list in the first embodiment of the present invention;

[0045] Figure 7 A detailed flow chart of S14 of the first embodiment of the present invention;

[0046] Figure 8 A detailed flow chart of S143 of the first embodiment of the present invention;

[0047] Fig. 9 A detailed flow chart of S15 of the first embodiment of the present invention;

[0048] Fig.10 It is a structural block diagram of a distribution network fault processing training device based on a decision support model according to a second embodiment of the present invention;

[0049] Fig.11 This is a schematic diagram of the internal structure of a computer according to another embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, and the well-known modules, units and their connections, links, communications or operations are not shown or described in detail. In addition, the described features, architectures or functions can be combined in any way in one or more embodiments. It should be understood by those skilled in the art that the various embodiments described below are only for illustration and not for limiting the scope of protection of the present invention. It can also be easily understood that the modules or units or processing methods in the various embodiments described herein and shown in the drawings can be combined and designed according to various different configurations. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] First embodiment:

[0052] Please refer to Figures 1 to 9 As shown, the embodiment of the present invention discloses a distribution network fault processing training method based on a decision support model, comprising:

[0053] S11, construct a distribution network fault recovery decision-making auxiliary model that includes real-time parameters of all topological elements.

[0054] As a specific solution but not a limitation, this step S11 is a pre-execution for building a distribution network fault recovery decision-making auxiliary model. Figure 2 As shown, this step S11 also includes S111-S113, wherein:

[0055] S111, constructing a basic topology model according to the topological relationship of the topological elements of the entire line, wherein the topological elements of the entire line include a main transformer of the main network, a busbar, an outgoing switch, a feeder switch of the distribution network, a distribution transformer and a switch, and dividing all the topological elements of the entire line into two element types, namely, points and edges, according to their opening and closing attributes;

[0056] S112, based on the basic topology model, introducing the real-time parameters of each of the full-circuit topology elements to construct a decision-making auxiliary model, wherein the real-time parameters include current, voltage, load and switch status;

[0057] The above-mentioned real-time parameters in this embodiment can be remote sensing data. Taking a distribution network containing one million devices, a real-time parameter collection frequency of once every 5 minutes, and a data size of each real-time parameter of 50 bytes as an example, the amount of telemetry data of the distribution network in one day is approximately 1000000*(24*60 / 5)*50B=14.4GB, and the amount of data of the distribution network in one month is 432GB.

[0058] S113, establishing the association operation between the GIS map model and the decision support model;

[0059] As a specific implementation of this embodiment but not limiting, this step S113 also includes S1131-S1133, wherein:

[0060] S1131, based on the management area, export the full distribution line model data of the GIS model.

[0061] By exporting the full amount of distribution line diagram data within the jurisdiction to the decision-making support model, the decision-making support model completes data initialization and further forms a complete power grid architecture.

[0062] S1132, the decision-making support model is based on the full distribution line diagram data, loads svg image technology through a web browser, uses JavaScript programming language to expand svg graphics, and adds animation effects to multi-line nodes and path nodes, thereby initializing and forming a visualization architecture of the power grid in the management area.

[0063] This step S1132 is used to achieve the technical purpose of intuitively feedback the energized state of the line through a power grid visualization architecture with animation effects.

[0064] At the same time, this step extends the GIS model SVG graphics, adds hotspots to the graphics elements of the GIS model network topology model, such as use, polyline and path nodes, and registers click events and right-click menus to the hotspots, so that the SVG graphics can be interactive and realize mouse control based on the Web browser page. At the same time, this step also includes collecting device information (metadata) in the SVG graphics, which is used to retrieve the location of the device in the current SVG graphics.

[0065] S1133, in response to user events, controlling the power grid visualization architecture to perform corresponding feedback operations, the user events including left-click / double-click, right-click, scroll wheel and drag operations, the feedback operations including calling up a menu, confirming, canceling, zooming in and out, and translating / rotating graphics.

[0066] Based on S1132, the decision support model in step S1133 can perform corresponding feedback operations in response to user events. For example, the user's right-click menu operation can process the parameters of the maintenance rehearsal function; when the user enters the device ID or device name and initiates a search, the visualization core adjusts the viewport based on the device information to display the device and highlights the device.

[0067] Please refer to Figure 3 As shown, after step S113, this embodiment further includes steps S114-S1110, wherein:

[0068] S114, constructing a weight element library, wherein the weight element library includes a plurality of weight element types.

[0069] The above-mentioned weight element library includes:

[0070] Whether it is a remotely controlled point for dispatching, the risk of voltage loss of the power outage equipment elements, the maximum current of the power outage equipment distribution network line in the month, the maximum current of the opposite ring network line in the month, the minimum value of the current carrying capacity of the opposite ring network line compared with the minimum current carrying capacity of the ring network loop, the number of users affected by the power outage equipment line, the number of users of the side distribution network line, the bus voltage difference where the line is located, the load rate of the main transformer and main network line and other main network equipment belonging to the opposite side line of the power transfer, the number of users affected by the power outage equipment bus voltage loss, the upper power supply can be traced back to the 10(20)kV ring network line that supplies power to the same 220kV substation, whether the line has completed phase sequence phase verification and trial closing test, whether it contains important sensitive users, the capacity-load ratio of transformers on both sides of the line, the secondary current of CT ratio on both sides of the line, the length of the power outage equipment distribution network line, the length of the opposite ring network line and the lines involved in the complaint work order.

[0071] S115, assigning a corresponding tendency label to each weight factor type.

[0072] The tendency labels are used to indicate the impact of each weight element type on the distribution network during fault handling. If necessary, different tendency labels can also belong to the same category or different categories to reflect the commonality of weight element types during fault handling.

[0073] S116, obtaining a training sample set, wherein the training sample set includes a plurality of distribution network fault topology models, each distribution network fault topology model includes a full-circuit topology element, simulation parameters of each full-circuit topology element, and a simulated predetermined transfer line.

[0074] In this embodiment, the ideal number of the above-mentioned multiple distribution network fault topology models is a large-scale data set, and the above-mentioned full-line topology elements include simulation parameters and technical parameters. The simulation parameters are used to simulate the real-time parameters of each full-line topology element when an accident occurs, and the technical parameters are used to reflect the technical indicators of the full-line topology element during static or normal operation. The simulated scheduled transfer line is the backup line of the node equipment of different branches when a simulated fault occurs, which corresponds to the scheduled transfer line of S123 described below.

[0075] S117, determining a plurality of simulated response plans based on the training sample set, and a plurality of standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics.

[0076] The simulated response plan, standard weight factor combination scheme and corresponding standard tendency characteristics of this step S117 can be set manually and used as an ideal result of the distribution network fault recovery decision-making auxiliary model.

[0077] S118, training the distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain multiple simulated response plans, multiple simulated weight factor combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics.

[0078] Please refer to Figure 4 As shown, in this step S118, steps S1181-S1187 are also included, wherein:

[0079] S1181, based on the distribution network fault topology model, a simulated directed acyclic graph topology diagram is generated through a distribution network fault recovery decision support model, wherein the simulated directed acyclic graph topology diagram includes a simulated predetermined transfer line.

[0080] S1182, based on the simulation parameters of all the topological elements of the entire road in the simulated directed acyclic graph topology diagram, determine the simulated faulty equipment, simulated fault range and simulated fault accident type.

[0081] S1183: Based on the device type of the affected device within the simulated fault range and the first decision mechanism, a corresponding weight element type is selected from a weight element library.

[0082] S1184, based on the priority and quantity of the affected devices within the fault range, as well as the technical parameters and simulation parameters of the affected devices, a weight assignment operation is performed for the corresponding weight element type according to the first decision mechanism.

[0083] S1185, determining a plurality of simulated response plans based on the simulated faulty equipment, the simulated fault scope, the simulated fault accident type, the simulated scheduled transfer line and the second decision-making mechanism.

[0084] S1186, based on multiple corresponding weight element types, weight values ​​and the first decision mechanism, generate multiple simulation weight element combination schemes corresponding to each simulation response plan.

[0085] S1187, generating a simulated tendency feature based on each corresponding simulated weight element combination scheme, the weight value of each weight element type in the combination scheme, the tendency label and the first decision mechanism.

[0086] In the above S1181-S1187, the first decision mechanism and the second decision mechanism can be respectively used as a whole, or can be respectively used as multiple decision units. The simulation tendency feature corresponds to the standard tendency feature, which is the scheme tendency corresponding to different simulation response plans and different weight factor combination schemes generated by the distribution network fault recovery decision auxiliary model during training, so as to facilitate managers to make decisions quickly.

[0087] S119, based on the tendency label, multiple standard weight element combination schemes and standard tendency characteristics, and corresponding simulation weight element combination schemes and simulation tendency characteristics, update the first decision mechanism of the distribution network fault recovery decision auxiliary model.

[0088] S1110, based on the distribution network fault topology model, multiple simulated response plans and corresponding standard response plans, update the second decision mechanism of the distribution network fault recovery decision auxiliary model.

[0089] S119-S1110 updates the first decision mechanism and the second decision mechanism by adopting different distribution network fault topology models, so that the decision support model can determine the weight factor combination scheme more accurately and efficiently, the simulated tendency characteristics can be as close to the standard tendency characteristics as possible, and the simulated response plan can be as close to the standard response plan as possible, providing managers with more reliable and accurate auxiliary information in the event of later power grid failures.

[0090] S12, establishing a first directed acyclic topology graph based on the decision support model.

[0091] Please refer to Figure 5 As shown, this step S12 also includes S121-S123, wherein:

[0092] S121, reading the XML file of the GIS map model based on the decision support model, converting the XML file into a document object, obtaining device elements therefrom, and defining and processing the attribute characteristics of the device elements through a programming language;

[0093] The class structure of the GIS model terminal includes at least the following five attributes: the dimension identifier id of the terminal list, the terminal identifier mrid, the serial number sequenceNumber that identifies the left and right sides of the device, the main device type conductingEquipment, and the mounted node connectivityNode. It also supports various protocols and custom protocols. For example, the conductingEquipment #SWITCH_00010102020012 represents a switch with a gisid of 00010102020012.

[0094] In this step, after the file processing unit extracts the terminal list object of the equipment element, it obtains two Terminal objects according to the main device type and the mounted node, which are defined as: nodeMap and equipmentMap.

[0095] S122, grouping, matching primitives and layout the device elements to construct an initial directed acyclic topology graph, wherein the initial directed acyclic topology graph includes node devices and edge devices.

[0096] For outgoing line switches and tie switches, a directed acyclic topology graph is constructed with the left and right sides as the starting positions. The layout includes storing the terminals and connection points of the GIS model as the edges of the directed acyclic topology graph in the order of the devices. Since each switch has a left and right side, 2N directed acyclic topology graphs will eventually be formed. This can solve the problem of uncertain power supply direction and the uncertainty of the front and back ends of the tie switch.

[0097] This step S122 can also further process the terminal list object, and its implementation function example is as follows: Figure 6 As shown: starting from the A equipment element, use gisid to obtain all its Terminals (the number is 1 or 2) from the equipmentMap, exclude Ta1 in the starting direction, use Ta2's connectivityNode, and obtain Tb1 and Tc1 mounted on the Node from the nodeMap. Since Tb1 contains the gisid of the B equipment element and Tc1 contains the gisid of the C equipment element, you only need to recursively perform the above steps and merge the results to complete the topological analysis of the entire graph.

[0098] When performing topological analysis, the device elements are recorded as vertices of the directed acyclic topological graph, for example, the device element of A is recorded as vertext_A, the device element of B is recorded as vertext_B, and the device element of C is recorded as vertex_C. The devices are sequentially stored in the edges of the directed acyclic topological graph, for example, Edge_AB represents an edge analyzed from device A to device B, and Edge_AC represents an edge analyzed from device A to device C. After the analysis is completed, the relationship between all device elements in the entire graph, such as the ancestor-descendant relationship, can be obtained.

[0099] On the other hand, the dispatcher customizes the vertex model of the directed acyclic topology graph and uses the device type and GISID to calculate the vertex hash value, which helps to quickly locate the device in various directed acyclic topology graphs.

[0100] S123, marking a predetermined transfer line in the initial directed acyclic topology graph to generate a first directed acyclic topology graph, wherein the predetermined transfer line is used to pre-connect node devices of different branches.

[0101] In this embodiment, the predetermined transfer lines are backup lines for node devices of different branches pre-connected in the initial directed acyclic topology graph, which are usually arranged in advance before an accident occurs. Usually, the predetermined transfer lines are arranged correspondingly for equipment with greater accident risks. The first directed acyclic topology graph aims to increase the stability and flexibility of the distribution network when an accident occurs.

[0102] S13, in response to the fault event, determining the faulty device based on the real-time parameters of the distribution network fault recovery decision support model.

[0103] The distribution network fault recovery decision-making auxiliary model monitors the real-time parameters of regional equipment in real time, generates fault events when the real-time parameters are abnormal, and quickly locks the faulty equipment to minimize the impact of power outages and speed up the recovery process.

[0104] S14, according to the faulty device, based on the distribution network fault recovery decision support model, analyzing the fault range in the first directed acyclic topology graph, and determining the fault accident type, wherein the fault accident type includes a single point fault and a multi-point fault.

[0105] In this embodiment, the distribution network fault recovery decision-making assistance model determines the type of fault accident, and based on whether the fault accident type is a single-point fault or a multi-point fault, combined with the rapid response plan corresponding to the predetermined transfer line, when some equipment with a higher accident risk fails, the rapid response plan can be used to efficiently complete fault handling and quickly restore power supply, thereby shortening or even eliminating the power outage time perception on the user side.

[0106] Please refer to Figure 7 As shown, this step S14 also includes S141-S144, wherein:

[0107] S141: If the faulty device is a single device, determine that the accident type is a single point failure.

[0108] When the faulty device is identified as a single device or line, it usually means that a single point of failure has occurred in the system. A single point of failure is a situation in which the failure of a single line or device in a distribution network system causes the entire system or part of the system to not function properly. In a power system, a single point of failure may include a tripped circuit breaker, a damaged transformer, or a shorted cable.

[0109] S142: If the faulty device is multiple devices, obtain the fault area boundary associated with the faulty device.

[0110] By analyzing the relationship between faulty devices and the network topology, the outermost circuit breaker or switch is determined as the boundary of the fault area. And the effectiveness of the determined boundary is verified by using a fault recovery decision-making auxiliary model based on real-time data. It should be noted that in this step S142, when the fault area boundaries of multiple devices intersect in the first directed acyclic topology graph, the accident type can be judged as a single point fault to simplify the difficulty of fault handling and ensure the efficiency of fault handling. If the fault area boundaries of multiple faulty devices do not intersect in the first directed acyclic topology graph, execute steps S143-S144.

[0111] S143, obtaining a predetermined transfer line associated with the boundary of each fault area, and analyzing the topological relationship between the predetermined transfer line and the faulty device and the affected device.

[0112] As a specific solution and not a limitation, please refer to Figure 8 As shown, this step S143 also includes S1431-S1432, wherein:

[0113] S1431, generating a second directed acyclic topology graph based on the fault isolation simulation of the faulty device and the affected device and after integrating one or more predetermined transfer lines;

[0114] In this step S1431, different fault isolation situations are simulated by adopting different predetermined transfer lines and combinations thereof, and different second directed acyclic topology graphs are generated based on the first directed acyclic topology graph for subsequent analysis.

[0115] S1432: Determine in the second directed acyclic topology graph whether fault area boundaries of multiple faulty devices intersect.

[0116] In this step S1432, it is determined whether the fault area boundaries of multiple faulty devices intersect in different second directed acyclic topology graphs. If so, it is determined that the current fault event has the conditions to be converted into a single point fault. If not, it is determined that the accident type corresponding to the current fault event is a multi-point fault.

[0117] S144: Based on the analysis result, determine whether the accident type is a single point failure or a multi-point failure.

[0118] In this step S144, the common scheme of the parameter of interest and the affected equipment can be used to determine whether the accident type is a single point failure or a multiple point failure. If the accident type is a multiple point failure, the multiple point failure can be treated as multiple single point failures.

[0119] S15, determining a matching rapid response plan and a weight factor combination scheme based on the type and parameters of the affected equipment within the fault range and the type of the fault accident.

[0120] The quick response plan of this embodiment corresponds to the predetermined transfer line, which is convenient for efficient fault handling, rapid power restoration, and shortening or even eliminating the power outage time perceived by the user. It should be noted that the quick response plan of this embodiment includes a pre-plan and a general plan. For single-point failures, the pre-plan includes but is not limited to the introduction of backup power supplies, or adjustment of power loads, etc. For multi-point failures, the pre-plan includes but is not limited to load transfer operations to ensure fault handling efficiency and rapid power restoration. Based on the above statement, the quick response plan includes:

[0121] Based on single point failure, introduce backup power supply or adjust power load;

[0122] Perform load transfer operations based on multiple point failures;

[0123] Based on a single point failure or multiple point failures, one or more associated scheduled transfer lines are enabled.

[0124] In this embodiment, please refer to Fig. 9 As shown, step S15 also includes S151-S154, wherein:

[0125] S151, based on the equipment type of the affected equipment within the fault range and different rapid response plans, select a corresponding weight factor type from a weight factor library;

[0126] S152, automatically generating a tendency feature based on a plurality of corresponding weight element types;

[0127] S153, performing a weight assignment operation for the corresponding weight element type based on the priority and quantity of the affected devices within the fault range, as well as the technical parameters and real-time parameters of the affected devices;

[0128] S154, establishing at least one set of weight element combination schemes according to each rapid response plan, wherein the weight element combination schemes include corresponding weight element types and weight values, and adding corresponding tendency features to each weight element combination scheme.

[0129] This embodiment can more conveniently match the weight element types in the same weight element combination scheme by establishing the association between the weight element types, so that different weight element combination schemes have more distinct tendency characteristics, so as to assist the administrator to make decisions quickly.

[0130] S16: Based on the weight factor combination scheme, a scoring operation is performed for each rapid response plan.

[0131] Corresponding to the above S51-S154, this step S16 is specifically as follows:

[0132] Based on multiple sets of weight factor combination schemes corresponding to each rapid response plan, a scoring operation is performed for each rapid response plan. At the same time, after the scoring operation is completed, the corresponding tendency characteristics are displayed so that the administrator can make a decision quickly based on the tendency characteristics.

[0133] As a preferred solution but not a limitation, this embodiment further includes S17-S18 after S16, wherein:

[0134] S17, determining a third directed acyclic topology graph based on the enabled one or more associated predetermined transfer lines;

[0135] S18, marking the faulty device and the isolated area in the third directed acyclic topology graph, and generating a target directed acyclic topology graph.

[0136] The third directed acyclic topology is one of the plurality of second directed acyclic topology graphs. Exemplarily, the third directed acyclic topology graph may be one of the second directed acyclic topology graphs with a higher priority for transfer. By marking the faulty equipment and the isolation area in the third directed acyclic topology graph, it is convenient to determine the maintenance plan for the faulty equipment and also to facilitate the subsequent overall analysis of the power supply plan.

[0137] The embodiment of the present invention constructs a weight element library, which includes multiple weight element types; assigns corresponding tendency labels to each weight element type; obtains a training sample set, which includes multiple distribution network fault topology models, each distribution network fault topology model includes a full-line topology element, simulation parameters of each full-line topology element and a simulated scheduled transfer line; determines multiple simulated response plans based on the training sample set, and multiple standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics; trains a distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain multiple simulated response plans, and multiple simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics; based on the tendency labels, multiple standard weight element combination schemes and standard tendency characteristics, and corresponding simulated weight element combination schemes and simulated tendency characteristics, updates the first decision mechanism of the distribution network fault recovery decision-making auxiliary model; based on the distribution network fault topology model, multiple simulated response plans and corresponding standard response plans, updates the second decision mechanism of the distribution network fault recovery decision-making auxiliary model. The embodiment of the present invention can provide a first decision-making mechanism and a second decision-making mechanism that are continuously updated and improved through training, so that the determination of the weight factor combination scheme is more accurate and efficient, the simulated tendency characteristics can be as close to the standard tendency characteristics as possible, and the simulated response plan can be as close to the standard response plan as possible, providing managers with more reliable and accurate auxiliary information in the event of a later power grid failure. At the same time, the embodiment of the present invention can improve the intelligence of the system by providing multiple sets of personalized power supply restoration schemes.

[0138] Second embodiment:

[0139] Please refer to Fig.10 As shown, the present invention also provides a distribution network fault processing training device 100 based on a decision support model, comprising:

[0140] A construction module 110 is used to construct a weight element library, wherein the weight element library includes a plurality of weight element types;

[0141] A matching module 120, for assigning a corresponding tendency label to each weight element type;

[0142] An acquisition module 130 is used to acquire a training sample set, wherein the training sample set includes a plurality of distribution network fault topology models, each of which includes a full-circuit topology element, simulation parameters of each full-circuit topology element, and a simulated predetermined transfer line;

[0143] A determination module 140 is used to determine a plurality of simulated response plans based on the training sample set, and a plurality of standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics;

[0144] A training module 150 is used to train the distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain multiple simulated response plans, multiple simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics;

[0145] A first updating module 160, configured to update a first decision mechanism of a distribution network fault recovery decision-making auxiliary model based on the tendency label, a plurality of standard weight element combination schemes and standard tendency characteristics, and corresponding simulation weight element combination schemes and simulation tendency characteristics;

[0146] The second updating module 170 is used to update the second decision mechanism of the distribution network fault recovery decision auxiliary model based on the distribution network fault topology model, multiple simulation response plans and corresponding standard response plans.

[0147] The modules of this embodiment are the same as the corresponding steps of the above method embodiment, and will not be repeated here.

[0148] The embodiment of the present invention constructs a weight element library, which includes multiple weight element types; assigns corresponding tendency labels to each weight element type; obtains a training sample set, which includes multiple distribution network fault topology models, each distribution network fault topology model includes a full-line topology element, simulation parameters of each full-line topology element and a simulated scheduled transfer line; determines multiple simulated response plans based on the training sample set, and multiple standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics; trains a distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain multiple simulated response plans, and multiple simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics; based on the tendency labels, multiple standard weight element combination schemes and standard tendency characteristics, and corresponding simulated weight element combination schemes and simulated tendency characteristics, updates the first decision mechanism of the distribution network fault recovery decision-making auxiliary model; based on the distribution network fault topology model, multiple simulated response plans and corresponding standard response plans, updates the second decision mechanism of the distribution network fault recovery decision-making auxiliary model. The embodiment of the present invention can provide a first decision-making mechanism and a second decision-making mechanism that are continuously updated and improved through training, so that the determination of the weight factor combination scheme is more accurate and efficient, the simulated tendency characteristics can be as close to the standard tendency characteristics as possible, and the simulated response plan can be as close to the standard response plan as possible, providing managers with more reliable and accurate auxiliary information in the event of a later power grid failure. At the same time, the embodiment of the present invention can improve the intelligence of the system by providing multiple sets of personalized power supply restoration schemes.

[0149] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0150] The present invention also provides a computer device, a distribution network fault processing training system based on a decision-making auxiliary model, and the distribution network fault processing training system based on a decision-making auxiliary model comprises: a distribution network fault processing training device based on a decision-making auxiliary model;

[0151] The distribution network fault processing training device of the decision support model is determined by the distribution network fault processing training method based on the decision support model shown in the second embodiment.

[0152] An embodiment of the present invention further provides a computer storage medium on which a computer program is stored. When the program is executed by a processor, the distribution network fault processing training method based on the decision support model as described in the above embodiments is implemented.

[0153] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it may include the processes of the embodiments of the distribution network fault processing training method based on the decision-making auxiliary model described above. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0154] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, RAM, ROM, magnetic disks or optical disks.

[0155] Corresponding to the above-mentioned computer storage medium, in one embodiment, a computer device is also provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a distribution network fault handling training method based on a decision support model as in the above-mentioned embodiments is implemented.

[0156] The computer device may be a terminal, and its internal structure diagram may be as follows: Fig.11 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a distribution network fault handling training method based on a decision-making auxiliary model is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0157] The embodiment of the present invention constructs a weight element library, which includes multiple weight element types; assigns corresponding tendency labels to each weight element type; obtains a training sample set, which includes multiple distribution network fault topology models, each distribution network fault topology model includes a full-line topology element, simulation parameters of each full-line topology element and a simulated scheduled transfer line; determines multiple simulated response plans based on the training sample set, and multiple standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics; trains a distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain multiple simulated response plans, and multiple simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics; based on the tendency labels, multiple standard weight element combination schemes and standard tendency characteristics, and corresponding simulated weight element combination schemes and simulated tendency characteristics, updates the first decision mechanism of the distribution network fault recovery decision-making auxiliary model; based on the distribution network fault topology model, multiple simulated response plans and corresponding standard response plans, updates the second decision mechanism of the distribution network fault recovery decision-making auxiliary model. The embodiment of the present invention can provide a first decision-making mechanism and a second decision-making mechanism that are continuously updated and improved through training, so that the determination of the weight factor combination scheme is more accurate and efficient, the simulated tendency characteristics can be as close to the standard tendency characteristics as possible, and the simulated response plan can be as close to the standard response plan as possible, providing managers with more reliable and accurate auxiliary information in the event of a later power grid failure. At the same time, the embodiment of the present invention can improve the intelligence of the system by providing multiple sets of personalized power supply restoration schemes.

[0158] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0159] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A distribution network fault processing training method based on a decision support model, characterized in that: include: Constructing a weight element library, wherein the weight element library includes a plurality of weight element types; Assign corresponding tendency labels to each weight factor type; Acquire a training sample set, wherein the training sample set includes a plurality of distribution network fault topology models, each distribution network fault topology model includes a full-circuit topology element, simulation parameters of each full-circuit topology element, and a simulated predetermined transfer line; Determining a plurality of simulated response plans based on the training sample set, and a plurality of standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics; Based on the training sample set, the distribution network fault recovery decision-making auxiliary model is trained to obtain a plurality of simulated response plans, a plurality of simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics; Based on the tendency label, multiple standard weight element combination schemes and standard tendency characteristics, and corresponding simulated weight element combination schemes and simulated tendency characteristics, updating the first decision mechanism of the distribution network fault recovery decision auxiliary model; Based on the distribution network fault topology model, multiple simulated response plans and corresponding standard response plans, the second decision mechanism of the distribution network fault recovery decision auxiliary model is updated.

2. The method according to claim 1, characterized in that The distribution network fault recovery decision-making auxiliary model is trained based on the training sample set to obtain multiple simulated response plans, multiple simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics, including: Based on the distribution network fault topology model, a simulated directed acyclic graph topology diagram is generated through a distribution network fault recovery decision-making auxiliary model, wherein the simulated directed acyclic graph topology diagram includes a simulated predetermined transfer line; Based on the simulation parameters of all the topological elements of the entire road in the simulated directed acyclic graph topology diagram, the simulated fault equipment, simulated fault range and simulated fault accident type are determined.

3. The method according to claim 2, characterized in that After determining the simulated fault equipment, simulated fault range and simulated fault accident type based on the simulation parameters of each full-road topology element in the simulated directed acyclic graph topology graph, the method includes: Based on the device type of the affected device within the simulated fault range and the first decision mechanism, a corresponding weight element type is selected from the weight element library; Based on the priority and quantity of the affected devices within the fault range, as well as the technical parameters and simulation parameters of the affected devices, a weight assignment operation is performed for the corresponding weight element type according to the first decision mechanism.

4. The method according to claim 3, characterized in that After performing a weight assignment operation for the corresponding weight element type based on the priority and quantity of the affected devices within the fault range, as well as the technical parameters and real-time parameters of the affected devices according to the first decision mechanism, the method includes: Determine multiple simulated response plans based on simulated fault equipment, simulated fault scope, simulated fault accident type, simulated scheduled transfer line and second decision-making mechanism; Based on a plurality of corresponding weight element types, weight values ​​and a first decision mechanism, a plurality of simulation weight element combination schemes corresponding to each simulation response plan are generated.

5. The method according to claim 4, characterized in that After generating a plurality of simulation weight element combination schemes corresponding to each simulation response plan based on a plurality of corresponding weight element types, weight values ​​and the first decision mechanism, the method includes: Based on the corresponding simulation weight element combination schemes, the weight values ​​of each weight element type in the combination schemes, the tendency labels and the first decision-making mechanism, a simulation tendency feature is generated.

6. The method according to claim 5, characterized in that The weight element library comprises: Whether it is a remotely controlled point for dispatching, the risk of voltage loss of the power outage equipment elements, the maximum current of the power outage equipment distribution network line in the month, the maximum current of the opposite ring network line in the month, the minimum value of the current carrying capacity of the opposite ring network line compared with the minimum current carrying capacity of the ring network loop, the number of users affected by the power outage equipment line, the number of users of the side distribution network line, the bus voltage difference where the line is located, the load rate of the main transformer and main network line and other main network equipment belonging to the opposite side line of the power transfer, the number of users affected by the power outage equipment bus voltage loss, the upper power supply can be traced back to the 10(20)kV ring network line that supplies power to the same 220kV substation, whether the line has completed phase sequence phase verification and trial closing test, whether it contains important sensitive users, the capacity-load ratio of transformers on both sides of the line, the secondary current of CT ratio on both sides of the line, the length of the power outage equipment distribution network line, the length of the opposite ring network line and the lines involved in the complaint work order.

7. A distribution network fault handling training device based on a decision support model, characterized in that: include: A construction module, used to construct a weight element library, wherein the weight element library includes multiple weight element types; A matching module is used to assign corresponding tendency labels to each weight factor type; An acquisition module is used to acquire a training sample set, wherein the training sample set includes a plurality of distribution network fault topology models, each of which includes a full-circuit topology element, simulation parameters of each full-circuit topology element, and a simulated predetermined transfer line; A determination module, configured to determine a plurality of simulated response plans based on the training sample set, and a plurality of standard weight element combination schemes corresponding to each simulated response plan, and corresponding standard tendency characteristics; A training module, used for training the distribution network fault recovery decision-making auxiliary model based on the training sample set to obtain a plurality of simulated response plans, a plurality of simulated weight element combination schemes corresponding to each simulated response plan, and corresponding simulated tendency characteristics; A first updating module, used to update a first decision mechanism of a distribution network fault recovery decision auxiliary model based on the tendency label, a plurality of standard weight element combination schemes and standard tendency characteristics, and corresponding simulation weight element combination schemes and simulation tendency characteristics; The second updating module is used to update the second decision mechanism of the distribution network fault recovery decision auxiliary model based on the distribution network fault topology model, multiple simulation response plans and corresponding standard response plans.

8. A power grid fault processing training system based on adaptive weight elements, characterized in that: The distribution network fault processing training system based on the decision-making auxiliary model comprises: a distribution network fault processing training device based on the decision-making auxiliary model; The distribution network fault processing training device based on a decision support model is determined by a distribution network fault processing training method based on a decision support model as shown in any one of claims 1-6.

9. A computer 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 computer program, the distribution network fault processing training method based on the decision support model as described in any one of claims 1 to 6 is implemented.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the distribution network fault processing training method based on the decision support model as described in any one of claims 1 to 6 is implemented.

Images