Visual monitoring analysis method and system for power failure

Abstract

The invention relates to a power failure visual monitoring analysis method and system. The method comprises the following steps: collecting and preprocessing operation data of regional power equipment; performing power failure behavior suspicion detection on the electricity meters in the area based on the area power equipment data; collecting geographic coordinates and topological structure data of power equipment in a region, and performing association matching on the topological structure data and the coordinate data of the power equipment to obtain a regional power structure topological graph; taking each electricity meter with power failure behavior suspicion as an initial node, and querying a corresponding superior topological structure in the regional power structure topological graph; performing fault prediction on all the power equipment in the superior topological structure of the ammeter with the suspected power failure behavior, and marking the power equipment of which the fault probability is greater than a threshold value; and performing fault source reasoning on the marked power equipment to obtain fault source power equipment, and marking a preset power failure influence range corresponding to the type of the fault source power equipment into the regional power structure topological graph.

Description

TECHNICAL FIELD

[0001] The present application relates to a power outage visual monitoring analysis method and system, belonging to the technical field of power monitoring. BACKGROUND

[0002] The prior art mainly relies on substation circuit breaker tripping signals or user telephone repair as the basis for power outage determination. Due to the complex level of distribution network, the average delay from fault occurrence to signal transmission to the monitoring center is more than 30 minutes. During this period, the operation and maintenance personnel cannot actively perceive the power loss state of the users at the end of the distribution network, resulting in a serious compression of the fault response window. Key data such as metering collection system (AMI), power grid topology data (GIS), and real-time weather information are in separate platforms, lacking a unified data fusion mechanism. When typhoons, thunderstorms and other disastrous weather occur, the operation and maintenance personnel need to manually cross-check the discrete data of multiple systems, which not only consumes a lot of time, but also results in insufficient fault positioning accuracy of less than 65% due to data timestamp deviation, seriously affecting the repair efficiency. The existing monitoring interface mainly uses tables or statistical charts to display power outage information, which cannot dynamically present the change trend of the power outage area in the spatial dimension. For cable trench fires, tree line discharges and other faults with spreading characteristics, the static display method is difficult to predict the diffusion direction and speed of the power outage range, resulting in a lag in the deployment of prevention and control measures.

[0003] The fault analysis process highly depends on the technical experience of the operation and maintenance personnel, and requires manual retrieval of historical data and step-by-step investigation of device status. Facing the complex distribution network topology, the average diagnosis time is more than 2 hours, and is prone to misjudgment due to human negligence. Passive response: relying on user repair or circuit breaker tripping signals, unable to actively predict power outage risks. Information island: metering data, topology data, and weather data are separated and cannot be correlated and analyzed. Lack of visualization: the power outage influence range is displayed in table form, lacking spatial dynamic rendering. Low diagnosis efficiency: manual fault chain investigation, time-consuming and prone to misjudgment. Especially when multiple associated devices are abnormal at the same time, the existing system cannot automatically build a fault conduction logic chain, making it difficult to quickly locate the core fault point. SUMMARY

[0004] In order to solve the problems existing in the prior art, the present application provides a power outage visual monitoring analysis method and system.

[0005] The technical solution of the present application is as follows: On the one hand, the present application provides a power outage visual monitoring analysis method, comprising the following steps: Collecting regional power equipment operation data; Suspected power outage behavior detection on the electric meter in the region based on the regional power equipment operation data; Collecting geographic coordinates of power equipment in the region and topology data, correlating and matching the topology data and the geographic coordinates of the power equipment data to obtain a regional power structure topology graph; Taking each power meter with a power outage behavior suspicion as a starting node, querying its corresponding superior topology structure in the regional power structure topology graph; Performing fault prediction on all power equipment in the superior topology structure of the power meter with a power outage behavior suspicion, and marking the power equipment with a fault probability greater than a threshold. Performing fault source reasoning on the marked power equipment to obtain a fault source power equipment, and marking a preset power outage influence range corresponding to the type of the fault source power equipment to the regional power structure topology graph.

[0006] Preferably, the regional power equipment operation data includes current data, voltage data, temperature data, load rate, and environmental data.

[0007] Preferably, the regional power equipment operation data is preprocessed, and the specific steps include data cleaning, data normalization, and data alignment.

[0008] Preferably, the specific steps for detecting the power meter in the region with a power outage behavior suspicion are as follows: For any power meter, extract the voltage of the adjacent power meter in the regional power topology structure, if the difference between the voltage of the power meter and the voltage of the adjacent power meter reaches a preset voltage difference threshold, and the voltage difference duration reaches a preset time threshold, it is determined that the power meter has a power outage behavior suspicion.

[0009] Preferably, the regional power structure topology graph includes a plurality of nodes, each node representing a power equipment, and after mapping the geographic coordinates of all power equipment to the spatial coordinate system of the regional power structure topology graph, arranging each node corresponding to the power equipment in the regional power structure topology graph according to the mapped coordinates, and connecting the nodes according to the connection relationship of all power equipment after the arrangement is completed. After mapping the regional map to the spatial coordinate system of the power structure topology graph, combining the mapped regional map and the regional power structure topology graph in layers to obtain a visual regional power structure topology graph.

[0010] Preferably, the specific steps for performing fault prediction on all power equipment in the superior topology structure of the power meter with a power outage behavior suspicion are as follows: Based on a long short-term memory network, a fault prediction model is constructed, the operation data of each power equipment is input to the fault prediction model, and the fault prediction model outputs the fault probability of each power equipment.

[0011] Preferably, the specific steps for performing fault source reasoning on the marked power equipment are as follows: Definition of conditional probability , indicating the marked power equipment The condition that the marked power equipment is the fault source The probability that the marked power equipment is the fault equipment , which is specifically shown in the following formula:

[0012] Wherein: The historical failure frequency of the marked power equipment ; The failure prediction probability of the marked power equipment ; The weight coefficient is represented by w; Based on the conditional probability, the overall likelihood probability is constructed, which is specifically shown in the following formula:

[0013] Wherein: The likelihood probability of the marked power equipment under the condition that the marked power equipment is the fault source ; Based on the Bayes formula, the posterior probability is calculated for each fault source assumption, which is specifically shown in the following formula:

[0014] Wherein: The posterior probability of the marked power equipment being the fault source ; The prediction failure probability of the marked power equipment ; The likelihood probability of the marked power equipment under the condition that the marked power equipment is the fault source ; The prediction failure probability of the marked power equipment ; The marked power equipment with the maximum posterior probability is regarded as the fault source power equipment of the power meter currently suspected of having power failure behavior.

[0015] On the other hand, the application also provides a power failure visual monitoring and analysis system, which comprises a data acquisition module, a power failure suspicion detection module, a power structure topology graph construction module, a topology structure query module, a fault equipment marking module and a fault source query module. The data acquisition module is used to acquire regional power equipment operation data. The power failure suspicion detection module is used to detect the power failure behavior suspicion of the power meter in the region based on the regional power equipment operation data. The power structure topology graph construction module is used for collecting geographic coordinates of power equipment in a region and topology structure data, and associating and matching the topology structure data and the geographic coordinates of the power equipment to obtain a regional power structure topology graph; The topology structure query module is used for taking each power meter with a power outage behavior suspicion as a starting node, and querying a corresponding superior topology structure thereof in the regional power structure topology graph; The fault equipment marking module is used for performing fault prediction on all power equipment in the superior topology structure of the power meter with the power outage behavior suspicion, and marking power equipment with a fault probability greater than a threshold value; The fault source query module is used for performing fault source reasoning on the marked power equipment to obtain a fault source power equipment, and marking a preset power outage influence range corresponding to a type of the fault source power equipment to the regional power structure topology graph.

[0016] In another aspect, the application further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method according to the application when executing the program.

[0017] In another aspect, the application further provides a computer readable storage medium, which stores a computer program, and the program is executable on a processor to implement the method according to the application.

[0018] The application has the following beneficial effects: 1. The application associates and matches topology structure data and geographic coordinate data of power equipment in a region, constructs a power structure topology graph, arranges device nodes according to mapping coordinates, and combines the device nodes with a regional map to realize integrated visualization of device topology relationship and geographic position, so that operation and maintenance personnel can intuitively view spatial distribution, connection relationship, and power outage influence range of the device on a unified interface, and the intuitive nature and efficiency of fault research and dispatching decision are greatly improved.

[0019] 2. After performing fault prediction on superior topology structure equipment and marking equipment with a fault probability greater than a threshold value, the application defines a conditional probability of "device historical fault frequency + fault prediction probability", constructs an overall likelihood function, finally solves a posterior probability of each candidate fault source by using a Bayes formula, and selects the equipment with the maximum posterior probability as a fault source, so that fault source positioning is more robust and has good interpretability. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 The application is a method flowchart. DETAILED DESCRIPTION

[0021] With reference to the drawings, the technical solutions in the embodiments of the present application will be clearly and completely described in order to make the technical solutions in the embodiments of the present application apparent to those skilled in the art. Obviously, the described embodiments are only a part rather than all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0022] It should be understood that the step numbers used herein are only for the convenience of description and are not limited to the execution sequence of the steps.

[0023] It should be understood that the terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application and the appended claims, unless otherwise clear from context, the singular forms "a", "an" and "the" are intended to include the plural forms as well.

[0024] The terms "comprise" and "include" indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0025] The term "and / or" means any combination of one or more of the associated listed terms and all possible combinations thereof, and includes these combinations.

[0026] Referring to Figure 1 In some embodiments, a power outage visualization monitoring analysis method is proposed, comprising the following steps: Collecting regional power equipment operation data and; Detecting power outage behavior suspicion of the electric meter in the region based on the regional power equipment operation data; Collecting the geographic coordinates of the power equipment in the region and the topological structure data, and correlating and matching the topological structure data and the geographic coordinates data of the power equipment to obtain a regional power structure topology map; Taking each electric meter with power outage behavior suspicion as a starting node, querying its corresponding superior topological structure in the regional power structure topology map; Performing fault prediction on all power equipment in the superior topological structure of the electric meter with power outage behavior suspicion, and marking the power equipment with a fault probability greater than a threshold; Performing fault source reasoning on the marked power equipment to obtain a fault source power equipment, and marking a preset power outage influence range corresponding to the type of the fault source power equipment to the regional power structure topology map.

[0027] In some embodiments, the regional power equipment operation data includes current data, voltage data, temperature data, load rate, and environmental data.

[0028] In some embodiments, the regional power device operation data is pre-processed, and the specific steps include data cleaning, data normalization, and data alignment.

[0029] In some embodiments, the specific steps for outage behavior suspicion detection of the electric meter in the region are as follows: For any electric meter, the voltage of the electric meter adjacent to it is extracted in the regional power topology structure. If the difference between the voltage of the electric meter and the voltage of the adjacent electric meter reaches the preset voltage difference threshold, and the voltage difference duration reaches the preset time threshold, it is determined that the electric meter has an outage behavior suspicion.

[0030] In a specific embodiment, for any electric meter, if the voltage difference between it and the adjacent electric meter in the regional power topology structure reaches 30% and the duration reaches 5 minutes, it is determined that the electric meter has an outage behavior suspicion.

[0031] In some embodiments, the regional power structure topology map includes a plurality of nodes, each node representing a power device. After mapping the geographic coordinates of all power devices to the spatial coordinate system of the regional power structure topology map, each node corresponding to a power device is arranged in the regional power structure topology map according to the mapped coordinates. After the arrangement is completed, the nodes are connected according to the connection relationship of all power devices. After mapping the regional map to the spatial coordinate system of the power structure topology map, the mapped regional map and the regional power structure topology map are combined into a layer to obtain a visual regional power structure topology map.

[0032] In a specific embodiment, the outage user density in each outage influence range is mapped to the HSV color space in the visual regional power structure topology map, and the saturation is positively correlated with the outage user density.

[0033] In a specific embodiment, the wind speed vector field is superimposed on the outage area in the visual regional power structure topology map.

[0034] In a specific embodiment, the topology structure takes the substation as the root node, the distribution line as the branch, and the distribution line includes device nodes such as transformers, circuit breakers, and disconnectors.

[0035] In a specific embodiment, for the power devices in the visual regional power structure topology map, the devices are divided into 500kV, 220kV, 110kV, etc. levels according to the voltage level. The display level is divided according to the power supply area. The display priority is set according to the outage influence range. Multi-level display is achieved by using hierarchical rendering technology.

[0036] In some embodiments, all power devices in the superior topology of the electricity meter suspected of causing power outage are fault predicted, and the specific steps are as follows: A fault prediction model is constructed based on a long short-term memory network, operation data of each power device is taken as an input of the fault prediction model, and the fault prediction model outputs a fault probability of each power device.

[0037] In a specific embodiment, a network structure of the fault prediction model is composed of an input layer, a hidden layer, and an output layer, and on this basis, hyperparameters such as a number of hidden layers, a number of neurons in the hidden layer, a learning rate, and a batch size are taken as optimization objects of a Bayesian optimization algorithm, the model is iterated on a verification set, a prior distribution is updated each time, and optimal hyperparameters are obtained.

[0038] In some embodiments, fault source reasoning is performed on the marked power device, and the specific steps are as follows: A conditional probability is defined , indicating that the marked power device is a condition for a fault source , and the probability that the marked power device is a fault device is specifically shown in the following formula:

[0039] Wherein: indicates a historical fault frequency of the marked power device ; indicates a fault prediction probability of the marked power device ; indicates a weight coefficient; An overall likelihood probability is constructed based on the conditional probability, and is specifically shown in the following formula:

[0040] Wherein: indicates a likelihood probability of the marked power device being a fault source under the condition ; indicates a set of marked power devices; Based on a Bayesian formula, a posterior probability of each fault source hypothesis is calculated, and is specifically shown in the following formula:

[0041] Wherein: indicates a posterior probability of the marked power device being a fault source; indicates a predicted fault probability of the marked power device . labeled power equipment a condition that is a fault source a likelihood probability under the condition; labeled power equipment a predicted fault probability of the labeled power equipment; the labeled power equipment with the maximum posterior probability is regarded as a fault source power equipment of the power meter currently suspected of having a power outage behavior.

[0042] In a specific embodiment, based on the regional power topology structure, the transfer paths available for use when each fault source power equipment fails are collected, the load margin of the transfer paths is calculated, and the transfer path with the highest load margin is selected as the fault transfer path.

[0043] In some embodiments, a power outage visualization monitoring and analysis system is provided, comprising a data acquisition module, a power outage suspicion detection module, a power structure topology graph construction module, a topology structure query module, a fault equipment labeling module, and a fault source query module. The data acquisition module is used to acquire regional power equipment operation data. The power outage suspicion detection module is used to detect power outage behavior suspicion of power meters in the region based on regional power equipment operation data. The power structure topology graph construction module is used to acquire regional power equipment geographic coordinates and topology structure data, and to associate and match the topology structure data and the power equipment geographic coordinate data to obtain a regional power structure topology graph. The topology structure query module is used to query the corresponding superior topology structure of each power meter suspected of having a power outage behavior as a starting node in the regional power structure topology graph. The fault equipment labeling module is used to perform fault prediction on all power equipment in the superior topology structure of the power meter suspected of having a power outage behavior, and to label the power equipment with a fault probability greater than a threshold. The fault source query module is used to perform fault source reasoning on the labeled power equipment to obtain a fault source power equipment, and to label a preset power outage influence range corresponding to the type of the fault source power equipment in the regional power structure topology graph.

[0044] In some embodiments, an electronic device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method according to any one of the embodiments of the present application when executing the program.

[0045] In some embodiments, a computer readable storage medium is provided, having a computer program stored thereon, which is executable by a processor to implement the method according to any one of the embodiments of the present application.

[0046] In the embodiments of the present application, “at least one” means one or more, and “multiple” means two or more. “And / or” describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the cases of A alone, A and B together, and B alone. Wherein A and B can be singular or plural. The character “ / ” generally represents an “or” relationship between the front and rear associated objects. “At least one of the following” and the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0047] Those of ordinary skill in the art can realize that the units and algorithm steps described in the embodiments disclosed herein can be realized in electronic hardware, computer software, and a combination of electronic hardware and computer software. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0048] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0049] In several embodiments provided by the present application, any function realized in the form of a software function unit and sold or used as an independent product can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that make contributions to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (Read-Only Memory; hereinafter referred to as: ROM), a random access memory (Random Access Memory; hereinafter referred to as: RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0050] The above only describes the embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation based on the content of the specification and drawings of the present application, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A power outage visualized monitoring analysis method, characterized by, The method comprises the following steps: Collecting regional power equipment operation data; Detecting power-off behavior suspicion of the electric meter in the region based on the regional power equipment operation data; Collecting the geographic coordinates of the power equipment in the region and the topological structure data, correlating and matching the topological structure data and the geographic coordinates data of the power equipment to obtain a regional power structure topological graph; Taking each electric meter with power-off behavior suspicion as a starting node, querying the corresponding superior topological structure thereof in the regional power structure topological graph; Performing fault prediction on all power equipment in the superior topological structure of the electric meter with power-off behavior suspicion, and marking the power equipment with a fault probability greater than a threshold value; Performing fault source reasoning on the marked power equipment to obtain a fault source power equipment, and marking a preset power-off influence range corresponding to the type of the fault source power equipment to the regional power structure topological graph.

2. The power outage visualization monitoring and analysis method of claim 1, wherein, The regional power equipment operation data comprises current data, voltage data, temperature data, load rate and environmental data.

3. The power outage visualization monitoring and analysis method of claim 1, wherein, The regional power equipment operation data is preprocessed, and the specific steps comprise data cleaning, data normalization and data alignment.

4. The power outage visualization monitoring and analysis method of claim 1, wherein, The specific steps for detecting power-off behavior suspicion of the electric meter in the region are as follows: For any electric meter, the voltage of the adjacent electric meter in the regional power topological structure is extracted, and if the difference between the voltage of the electric meter and the voltage of the adjacent electric meter reaches a preset voltage difference threshold value and the voltage difference duration reaches a preset time threshold value, it is judged that the electric meter has power-off behavior suspicion.

5. The power outage visualization monitoring and analysis method of claim 1, wherein, The regional power structure topological graph comprises a plurality of nodes, each node representing a power equipment, and after mapping the geographic coordinates of all power equipment to the spatial coordinate system of the regional power structure topological graph, the nodes corresponding to each power equipment are arranged according to the mapped coordinates in the regional power structure topological graph, and after the arrangement is completed, the nodes are connected according to the connection relationship of all power equipment; After mapping the regional map to the spatial coordinate system of the power structure topological graph, the mapped regional map and the regional power structure topological graph are combined in layers to obtain a visual regional power structure topological graph.

6. The power outage visualization monitoring and analysis method of claim 1, wherein, The specific steps for performing fault prediction on all power equipment in the superior topological structure of the electric meter with power-off behavior suspicion are as follows: A fault prediction model is constructed based on a long short-term memory network, the operation data of each power equipment is input into the fault prediction model, and the fault prediction model outputs the fault probability of each power equipment.

7. The power outage visualization monitoring and analysis method of claim 1, wherein, The specific steps for performing fault source reasoning on the marked power equipment are as follows: Defining conditional probabilities , indicating the marked power device The condition that is the source of the fault , indicating the marked power device The probability that the marked device is the source of the fault is given by the following equation: wherein: represents the historical failure frequency of the marked power device ; represents the failure prediction probability of the marked power device ; represents the weight coefficient; An overall likelihood probability is constructed based on conditional probability, and the specific formula is as follows: wherein: a marked power device a condition that is a source of the fault a likelihood probability under the condition; a set of marked power devices; The posterior probability of each fault source hypothesis is calculated based on the Bayes formula, and the specific formula is as follows: wherein: represents a labeled power device is a posterior probability that the fault source; represents a labeled power device is a predicted probability of failure; represents a labeled power device is a likelihood probability that the fault source; under the condition represents a labeled power device is a predicted probability of failure; The marked power equipment with the maximum posterior probability is regarded as the fault source power equipment of the current electric meter with power-off behavior suspicion.

8. A power outage visualized monitoring analysis system, characterized by The system comprises a data collection module, a power-off suspicion detection module, a power structure topological graph construction module, a topological structure query module, a fault equipment marking module and a fault source query module; The data collection module is used to collect regional power equipment operation data; The power-off suspicion detection module is used to detect power-off behavior suspicion of the electric meter in the region based on the regional power equipment operation data; The power structure topology graph construction module is configured to collect geographic coordinates of power equipment in a region and topology structure data, associate and match the topology structure data and the geographic coordinates of the power equipment, and obtain a regional power structure topology graph; The topology structure query module is configured to take each power meter with a power outage behavior suspicion as a starting node, and query a corresponding superior topology structure in the regional power structure topology graph; The faulty equipment marking module is configured to perform fault prediction on all power equipment in the superior topology structure of the power meter with the power outage behavior suspicion, and mark power equipment with a fault probability greater than a threshold value; The fault source query module is configured to perform fault source reasoning on the marked power equipment to obtain a fault source power equipment, and mark a preset power outage influence range corresponding to a type of the fault source power equipment to the regional power structure topology graph.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method of any one of claims 1 to 7.

Images