Power failure fault research and determination positioning method and system for low-voltage transformer area

Abstract

The invention discloses a low-voltage transformer area-oriented power failure fault research and positioning method and system, and relates to the technical field of electric power, and the method comprises the steps: collecting power failure event data of a low-voltage transformer area in real time; performing multi-dimensional feature analysis on the power failure event data, and constructing a transformer area topological relation network according to space-time correlation features; generating a fault feature vector based on the space-time correlation feature and the transformer area topological relation network, performing mode matching according to the fault feature vector, and determining a fault matching result; and carrying out positioning verification on the low-voltage transformer area according to a fault matching result, and generating a fault positioning parameter. Therefore, the technical effects of quickly and efficiently realizing the power failure fault diagnosis and positioning in the low-voltage transformer area and improving the positioning accuracy and reliability are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power, and particularly to a method and system for judging and locating power outage faults for low-voltage power distribution areas. Background Art

[0002] The low-voltage power distribution area is an important part of the power distribution network in the power system, and its operating state directly affects the power consumption experience of users. In actual operation, power outage events may occur in the low-voltage power distribution area due to equipment failures, line aging, external interference, etc.

[0003] In the prior art, the judgment method of power outage faults usually only relies on single spatio-temporal features or simple topological relationships. For example, fault judgment is only carried out through time series data or geographical location information, but the equipment connection relationship and spatio-temporal correlation features in the power distribution area are not effectively combined, resulting in insufficient fault location accuracy, low efficiency and easy omission of key information, and being unable to adapt to the complex and changeable operating environment of the low-voltage power distribution area. Summary of the Invention

[0004] The present invention provides a method and system for judging and locating power outage faults for low-voltage power distribution areas to solve the technical problems of low fault location efficiency, low accuracy and poor reliability in the prior art, and to achieve the technical effects of quickly and efficiently diagnosing and locating power outage faults in the low-voltage power distribution area and improving the location accuracy and reliability.

[0005] In a first aspect, the present invention provides a method for judging and locating power outage faults for low-voltage power distribution areas, wherein the method for judging and locating power outage faults for low-voltage power distribution areas includes:

[0006] Real-time collection of power outage event data of the low-voltage power distribution area.

[0007] Perform multi-dimensional feature analysis on the power outage event data, and construct a topological relationship network of the power distribution area according to spatio-temporal correlation features.

[0008] Generate a fault feature vector based on the spatio-temporal correlation features and the topological relationship network of the power distribution area, and perform pattern matching according to the fault feature vector to determine the fault matching result.

[0009] Perform location verification on the low-voltage power distribution area according to the fault matching result, and generate fault location parameters.

[0010] In a feasible implementation manner, the method for real-time collection of power outage event data of the low-voltage power distribution area includes:

[0011] Extract data from the low-voltage power distribution area through the intelligent meter terminal to obtain voltage and current waveform data.

[0012] Receive load mutation data of the distribution transformer monitoring terminal according to the voltage and current waveform data.

[0013] Retrieve the power consumption information dataset, match the power consumption information dataset with the load mutation data, and determine multiple power loss information.

[0014] Perform timestamp alignment processing on the multiple power loss information to determine the power outage event data.

[0015] In a feasible implementation, perform multi-dimensional feature analysis on the power outage event data, and construct a substation area topology relationship network according to the spatio-temporal association features. The method includes:

[0016] Introduce the geographical information data of the low-voltage substation area for analysis, and construct a device connection relationship graph according to the geographical analysis data combined with the operation, distribution, and dispatching ledger information.

[0017] Generate a spatial topology tree structure according to the installation location of the smart meter and the physical coordinates of the line branch box.

[0018] Annotate the spatial topology tree structure according to the device connection relationship graph to construct the substation area topology relationship network.

[0019] In a feasible implementation, generate a fault feature vector based on the spatio-temporal association features and the substation area topology relationship network. The method includes:

[0020] Perform normalization processing on the spatio-temporal association features to generate normalized feature values.

[0021] Perform weight calculation based on the substation area topology relationship network combined with the normalized feature values to generate multi-dimensional feature weight distribution parameters.

[0022] Perform matrix splicing on the normalized feature values and the multi-dimensional feature weight distribution parameters to generate the fault feature vector.

[0023] In a feasible implementation, perform pattern matching according to the fault feature vector to determine the fault matching result. The method includes:

[0024] Retrieve multiple historical fault case data, perform power outage fault analysis based on the multiple historical fault case data, and formulate a fault mode library.

[0025] Use the fault feature vector as an index to perform retrieval calculation on the fault mode library to obtain a fault mode similarity coefficient.

[0026] Set a fault threshold according to the multiple historical fault case data. When the fault mode similarity coefficient is greater than or equal to the fault threshold, trigger an associated fault instruction.

[0027] Trace back the associated fault instruction to the fault mode library for pattern matching to determine the fault matching result.

[0028] In a feasible implementation manner, the low-voltage power distribution area is located and verified according to the fault matching result, and fault location parameters are generated. The method includes:

[0029] Data conflict detection is performed on the low-voltage power distribution area based on the associated fault instruction, and equipment grading identification is performed on the low-voltage power distribution area according to the conflict detection result, and a grading result is generated. The conflict detection result includes a candidate fault point set.

[0030] Weighted analysis is performed on the grading result according to the fault matching result to obtain an equipment weight coefficient, and probability sorting is performed on the candidate fault point set according to the equipment weight coefficient to generate a fault point sequence.

[0031] Verification is performed based on the fault point sequence in combination with the grading result to determine the fault location parameters.

[0032] In a feasible implementation manner, after the fault location parameters are generated, the method includes:

[0033] Impact analysis is performed based on the fault point sequence to determine the power outage impact range, and a repair priority is formulated according to the power outage impact range.

[0034] The repair priority is matched with the fault point sequence to generate a fault repair path.

[0035] The fault location parameters are traversed and checked for duplication according to the fault repair path to formulate a power outage fault repair strategy.

[0036] In a second aspect, the present invention also provides a power outage fault judgment and location system for a low-voltage power distribution area. Among them, the power outage fault judgment and location system for a low-voltage power distribution area includes:

[0037] An event data acquisition module, configured to collect power outage event data of the low-voltage power distribution area in real time.

[0038] A topology relationship construction module, configured to perform multi-dimensional feature analysis on the power outage event data, and construct a power distribution area topology relationship network according to the spatio-temporal association features.

[0039] A feature vector generation and matching module, configured to generate a fault feature vector based on the spatio-temporal association features and the power distribution area topology relationship network, and perform pattern matching according to the fault feature vector to determine a fault matching result.

[0040] A location verification module, configured to perform location verification on the low-voltage power distribution area according to the fault matching result, and generate fault location parameters.

[0041] The present invention discloses a method and system for judging and locating power outage faults in low-voltage power distribution areas, including: collecting power outage event data of low-voltage power distribution areas in real time; performing multi-dimensional feature analysis on the power outage event data, and constructing a topological relationship network of the power distribution area according to spatio-temporal correlation features; generating a fault feature vector based on the spatio-temporal correlation features and the topological relationship network of the power distribution area, performing pattern matching according to the fault feature vector, and determining the fault matching result; performing positioning verification on the low-voltage power distribution area according to the fault matching result, and generating fault positioning parameters. The method and system for judging and locating power outage faults in low-voltage power distribution areas disclosed by the present invention solve the technical problems of low fault location efficiency, low accuracy, and poor reliability, and achieve the technical effects of quickly and efficiently diagnosing and locating power outage faults in low-voltage power distribution areas, and improving the positioning accuracy and reliability. Description of the Drawings

[0042] Figure 1 It is a flow chart of the method for judging and locating power outage faults in low-voltage power distribution areas according to the present invention.

[0043] Figure 2 It is a structural diagram of the system for judging and locating power outage faults in low-voltage power distribution areas according to the present invention.

[0044] Description of the reference numerals: Event data acquisition module 11, topological relationship construction module 12, feature vector generation and matching module 13, positioning verification module 14. Detailed Embodiments

[0045] The above technical solutions will be described in detail below in combination with the drawings in the specification and specific embodiments to better understand the above technical solutions. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments for explaining the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention. In addition, it should be noted that only the parts related to the present invention rather than all are shown in the drawings for the convenience of description.

[0046] Embodiment 1, as Figure 1 It is a flow chart of the method for judging and locating power outage faults in low-voltage power distribution areas according to the present invention. Among them, the method for judging and locating power outage faults in low-voltage power distribution areas includes:

[0047] S100: Collect power outage event data of low-voltage power distribution areas in real time.

[0048] Specifically, the power outage event data are various data reflecting the historical power outage situations in the low-voltage power distribution area. Exemplarily, the power outage event data include voltage sag sequences, abnormal current waveforms, and user repair information. By collecting the power outage event data of the low-voltage power distribution area in real time, the timeliness and accuracy of the fault judgment and location method can be ensured. The data collected in real time can most accurately reflect the recent power outage situations in the power distribution area, providing a reliable basis for subsequent multi-dimensional feature analysis and fault location.

[0049] In some embodiments, to collect the power outage event data of the low-voltage power distribution area in real time, the method includes:

[0050] Extract data from the low-voltage power distribution area through the smart meter terminal to obtain voltage and current waveform data; receive the load mutation data of the distribution transformer monitoring terminal according to the voltage and current waveform data; retrieve the power consumption information data set, and match it with the load mutation data according to the power consumption information data set to determine multiple power loss information; perform timestamp alignment processing on the multiple power loss information to determine the power outage event data.

[0051] Specifically, the smart meter terminal is a smart device installed at the user end for real-time monitoring and recording of meter data, including information such as voltage and current. Among them, the voltage and current waveform data are curve data reflecting the changes of voltage and current over time.

[0052] Specifically, the distribution transformer monitoring terminal is a monitoring device installed at the distribution transformer for real-time monitoring of the load change situation of the transformer in the low-voltage power distribution area; the load mutation data of the distribution transformer monitoring terminal are data reflecting the sudden change of the load of the distribution transformer, including information related to power outage events.

[0053] Specifically, the power consumption information data set is a data set of the power consumption information of multiple users in the target low-voltage power distribution area. Exemplarily, it includes the power consumption and power consumption time of users, etc.; the power loss information refers to the information that a certain user or device loses power supply. For example, at the same moment, the power consumption information data and the load mutation data both fluctuate in the same direction (such as the power consumption and the load both decrease).

[0054] Specifically, first, data extraction is performed on the low-voltage power distribution area through the intelligent electricity meter terminal to obtain voltage and current waveform data, which can reflect the voltage and current changes at the user end. Then, according to these waveform data, the load mutation data of the corresponding distribution transformer monitoring terminal is received. Next, the electricity consumption information data set is retrieved and matched with the load mutation data to determine multiple power outage information. Finally, these power outage information are subjected to timestamp alignment processing to ensure the time consistency of the data, thereby determining accurate power outage event data. Exemplarily, in a low-voltage power distribution area containing multiple users, the intelligent electricity meter terminal can monitor the voltage and current changes of each user in real time. When a sudden voltage drop of a certain user is detected, this change is recorded and used as part of the power outage event data.

[0055] Through the above process, real-time collection of power outage event data in the low-voltage power distribution area can ensure the timeliness and accuracy of the fault judgment and location method. Among them, through the collaborative work of the intelligent electricity meter terminal, the distribution transformer monitoring terminal, and the electricity consumption information data set, all characteristics of the power outage event can be comprehensively captured, ensuring the integrity and accuracy of the data, and providing a reliable basis for subsequent multi-dimensional feature analysis and fault location.

[0056] S200: Perform multi-dimensional feature analysis on the power outage event data, and construct a topological relationship network of the power distribution area according to the spatio-temporal correlation features.

[0057] Specifically, perform correlation analysis based on feature extraction on the power outage event data from different feature dimensions (such as time, space, device status, etc.), identify the correlations between different intelligent electricity meter terminals (corresponding to different users), and form a topological relationship network of the power distribution area. This topological relationship network of the power distribution area is the connection relationship network between devices in the low-voltage power distribution area, usually represented in the form of a graph; among them, the nodes in the network represent devices, and the edges represent the connection relationships between devices.

[0058] Specifically, the spatio-temporal correlation features refer to the key data features that reflect the distribution laws of fault events in time and space, such as the moment when the fault occurs, the duration, the relative position of the fault point, and the propagation path.

[0059] Through multi-dimensional feature analysis and the construction of the topological relationship network of the power distribution area, the accuracy and efficiency of fault location can be significantly improved; among them, multi-dimensional feature analysis can comprehensively capture the spatio-temporal correlation characteristics of the power outage event, providing rich feature data for the subsequent generation of fault feature vectors; the topological relationship network of the power distribution area provides a structured analysis framework for fault location, enabling the fault point to be quickly located in the network.

[0060] In some embodiments, performing multi-dimensional feature analysis on the power outage event data and constructing a topological relationship network of the power distribution area according to the spatio-temporal correlation features, the method includes:

[0061] Introduce the geographical information data of the low-voltage substation area for analysis, and construct a device connection relationship graph based on the geographical analysis data in combination with the operation, distribution, and dispatching ledger information; generate a spatial topological tree structure according to the physical coordinates of the smart meter installation location and the line branch box; annotate the spatial topological tree structure according to the device connection relationship graph to construct the topological relationship network of the substation area.

[0062] Specifically, the topological relationship network of the substation area is a graph structure model representing the devices and their connection methods in the substation area. The operation, distribution, and dispatching ledger refers to the account books of power supply enterprises regarding power marketing, distribution management, and dispatching equipment and user information, which are used to record the technical parameters and wiring relationships of facilities such as transformers, switches, and lines. The device connection relationship graph is a graph data structure constructed based on the ledger data, which is used to display the connection and logical relationships between devices. The spatial topological tree structure is a tree structure mapped based on physical coordinates and device connection relationships, which is used to represent the hierarchy and distribution of the substation area in space.

[0063] Specifically, first, introduce the geographical information data of the low-voltage substation area, such as the longitude and latitude or address coordinates of transformers, branch boxes, switches, and meters, perform geographical analysis in combination with the GIS system, and match the information after geographical analysis with the operation, distribution, and dispatching ledger data of the power grid enterprise to construct a device connection relationship graph, initially depicting the logical and electrical connection structure between devices; then, according to the physical installation location of each smart meter and the location of the branch box of its affiliated line, generate the corresponding spatial topological tree structure through the spatial position mapping algorithm. The root node of the tree is usually the transformer, the branch nodes are the branch boxes, and the leaf nodes are the terminal meters to display the distribution of devices in space; finally, use the constructed device connection relationship graph to annotate the spatial topological structure, that is, label each node in the tree with device ID, type, connection status, etc., so as to construct a topological relationship network of the substation area including spatial hierarchical relationships and electrical connection logics, providing a structured basic support for subsequent power outage location, fault attribution, and operation and maintenance optimization, and helping to quickly locate the fault point in the network.

[0064] S300: Generate a fault feature vector based on the spatio-temporal correlation feature and the topological relationship network of the substation area, perform pattern matching according to the fault feature vector, and determine the fault matching result.

[0065] Specifically, the fault feature vector is a vectorized data expression integrated from multi-dimensional features (such as current, voltage, temperature, and spatio-temporal parameters); through this fault feature vector, it is convenient to perform pattern matching, that is, through algorithms, compare and match in a known feature template library to determine the fault mode corresponding to the current fault feature vector.

[0066] Exemplarily, in a power grid substation area, when there is a sudden sharp drop in current in a certain area, accompanied by a fault voltage fluctuation, and according to the topological structure analysis, the fault may affect the connected substations. At this time, the corresponding features are abstracted into vectors and matched with similar low-current and voltage abnormal diffusion patterns in the historical records to finally determine the fault matching result, so as to quickly locate the problem area and take maintenance measures.

[0067] In some embodiments, a fault feature vector is generated based on the spatio-temporal correlation feature and the substation area topological relationship network. The method includes:

[0068] Normalize the spatio-temporal correlation feature to generate a normalized feature value; calculate weights based on the substation area topological relationship network in combination with the normalized feature value to generate a multi-dimensional feature weight distribution parameter; splice the normalized feature value and the multi-dimensional feature weight distribution parameter in a matrix to generate the fault feature vector.

[0069] Specifically, first, normalize the collected spatio-temporal correlation features (such as power outage duration, number of affected users, geographical density, etc.) to make them on the same numerical scale, thereby eliminating the influence of inconsistent dimensions. Then, based on the substation area topological relationship network, combine the normalized feature values with the network structure information, and perform feature propagation and weighting according to the connection relationship of each node in the network (such as the number of neighbors, depth, propagation path, etc.) to calculate the influence degree of each dimension feature on the fault event in the whole network, so as to generate the weight distribution parameter of the multi-dimensional features. For example, a central node of a certain substation area in the network often has a greater impact on fault propagation, and its corresponding spatio-temporal features may be given higher weights. Finally, splice the above-mentioned normalized feature values and feature weight distribution parameters in a specific dimension to form a fault feature vector with unified expression. This vector combines spatio-temporal information, electrical structure weights, and feature normalization information for input or analysis and processing in subsequent models.

[0070] By combining the normalized spatio-temporal features and the structured topological weight distribution, the generated fault feature vector can more comprehensively and accurately depict the essential characteristics of a power outage event. This vector not only retains the local features of the event but also captures the global context information through the topological structure, which helps to improve the accuracy and robustness of subsequent fault recognition, location, classification and other models.

[0071] In some embodiments, pattern matching is performed according to the fault feature vector to determine the fault matching result. The method includes:

[0072] Retrieve multiple historical fault case data, conduct power outage fault analysis based on the multiple historical fault case data, and formulate a fault mode library; use the fault feature vector as an index to perform retrieval calculations on the fault mode library to obtain a fault mode similarity coefficient; set a fault threshold according to the multiple historical fault case data, and when the fault mode similarity coefficient is greater than or equal to the fault threshold, trigger an associated fault instruction; trace back the associated fault instruction to the fault mode library for pattern matching to determine the fault matching result.

[0073] Specifically, historical fault case data refers to the data records of past power outage fault events, including fault characteristics, location results, etc.; the fault mode library is a database established through statistical analysis and pattern recognition based on historical fault case data, used to store various fault modes and their corresponding characteristics (which can be represented by feature vectors).

[0074] Specifically, it refers to the quantitative representation of the similarity degree between the current fault feature vector and the feature vector of a certain fault mode in the fault mode library. For example, 0 represents completely dissimilar, and 1 represents completely similar; the fault threshold is a preset similarity coefficient threshold used to determine whether to trigger an associated fault instruction. In other words, exceeding this fault threshold can be considered that the fault feature vector corresponds to a fault with a specific mode.

[0075] Specifically, extract data from the monitoring system, inspection records, and fault reports, and ensure the integrity and consistency of the data. Then, based on the large amount of historical data collected, specifically conduct power outage fault analysis: for example, the occurrence frequency, fault duration, equipment anomaly indicators, and direct or indirect causes of power outages in different substations and different time periods can be statistically analyzed, and then representative feature patterns during fault occurrence can be extracted through data mining techniques (such as clustering analysis, association rule mining, or machine learning algorithms). Subsequently, combining these statistical and analysis results, summarize the typical fault characteristics, formulate fault modes that can represent different power outage fault situations, and save these modes in the mode library for future reference in fault identification, response, and prevention measures.

[0076] Specifically, use the generated fault feature vector as an index to perform retrieval calculations in the fault mode library to obtain the similarity coefficients of the fault feature vector to different fault modes, and set a fault threshold according to the historical fault case data. When the similarity coefficient is greater than or equal to this threshold, trigger an associated fault instruction, that is, locate the faulty equipment (coarse positioning). Furthermore, trace back the associated fault instruction to the fault mode library for pattern matching to determine the specific fault matching result. Among them, the fault matching result includes the fault type and the initial location range.

[0077] Through the above process, the fault matching results can be determined quickly and accurately, significantly improving the efficiency and accuracy of fault location. Among them, the pattern matching process uses the fault pattern library established by historical fault data to achieve judgment based on a large number of actual cases, thereby improving the accuracy and reliability of the judgment.

[0078] S400: Perform location verification on the low voltage substation according to the fault matching result and generate fault location parameters.

[0079] Specifically, the initially determined fault location is verified to ensure the accuracy of the fault location, wherein the fault location parameters are parameters used to accurately locate the fault location, usually including the coordinates of the fault point, the impact range, etc. Preferably, the fault location parameters can be characterized as a fault path backtracking chain, including the fault source and the corresponding multiple affected devices and impact paths.

[0080] In some embodiments, the low voltage area is located and verified according to the fault matching result to generate fault location parameters, and the method includes:

[0081] Based on the associated fault instructions, data conflict detection is performed on the low-voltage substation, and equipment classification is identified for the low-voltage substation according to the conflict detection result to generate a classification result, wherein the conflict detection result includes a set of candidate fault points; weighted analysis is performed on the classification result according to the fault matching result to obtain equipment weight coefficients, and probability sorting of the candidate fault point set is performed according to the equipment weight coefficients to generate a fault point sequence; verification is performed based on the fault point sequence in combination with the classification result to determine the fault location parameters.

[0082] Specifically, according to the associated fault instructions, multiple possible faulty devices (coarse positioning results) in the low-voltage area are traversed for data conflict detection, that is, the data consistency between different data sources or devices is checked to verify the accuracy of fault location. For example, if a branch box reports a power outage, but the smart meter downstream still shows power, this may indicate a data conflict and requires further investigation and storage in the candidate fault point set.

[0083] Furthermore, according to the results of conflict detection, the equipment in the substation is graded, such as being marked as first-level key equipment (such as main power supply equipment), second-level associated equipment (such as branch nodes), etc., and a device map with grade identification is formed; then, the candidate fault point set is probabilistically sorted according to the equipment weight coefficient to generate a fault point sequence. For example, equipment with higher weights ranks higher in the sequence and is verified first; finally, based on the fault point sequence combined with the grading results, the fault location parameters are determined. For example, if multiple high-weight devices point to a branch box at the same time, the branch box can be determined as a fault point, and the corresponding fault location parameters can be generated, such as the coordinates of the fault point and the impact range.

[0084] Through the above process, the accuracy and reliability of fault location can be significantly improved. Among them, data conflict detection ensures the consistency and accuracy of the data used for location, reducing the possibility of misjudgment; device classification identification and weighted analysis help to prioritize the handling of critical devices, thus improving the efficiency of fault location.

[0085] In some embodiments, after generating the fault location parameters, the method includes:

[0086] Based on the fault point sequence, conduct impact analysis to determine the power outage impact range, and formulate repair priorities according to the power outage impact range; match the repair priorities with the fault point sequence to generate a fault repair path; traverse and check the fault location parameters according to the fault repair path to formulate a power outage fault repair strategy.

[0087] Specifically, first, based on the existing fault point sequence (that is, a series of key fault nodes determined according to the fault location parameters), analyze the association and mutual influence between each fault point, identify the radiation or chain effect after a single fault point is activated, so as to determine the impact range of the entire power outage event. Then, according to this impact range, sort each power outage affected area or single fault point according to power outage impact degree, recovery difficulty, user importance and other key indicators, and formulate repair priorities, so as to ensure that high-priority areas or fault points will be given more urgent repair requirements.

[0088] Further, after obtaining the repair priorities of each fault point, match the repair priorities with the fault point sequence to form a reasonable fault repair path. Optionally, the matching process is implemented by integrating a scheduling algorithm or a path planning algorithm, so as to reasonably arrange maintenance personnel to first handle high-priority areas and then gradually extend to low-priority areas under the premise of meeting the requirement of restoring power supply as soon as possible, improving the response efficiency.

[0089] Further, according to the generated fault repair path, traverse and check the previously generated fault location parameters to ensure that each fault point is only processed once in the repair plan, avoid the impact caused by data redundancy or repeated analysis, and further improve and form the final power outage fault repair strategy. This strategy can not only accurately guide relevant technicians to check and repair faults one by one, but also be used as a support for subsequent automated scheduling and emergency response decision-making.

[0090] In summary, the power outage fault judgment and location method for low-voltage power distribution areas provided by the present invention has the following technical effects:

[0091] By collecting power outage event data of low-voltage power distribution areas in real time; performing multi-dimensional feature analysis on the power outage event data, constructing a topological relationship network of the power distribution area according to spatio-temporal correlation features; generating a fault feature vector based on the spatio-temporal correlation features and the topological relationship network of the power distribution area, performing pattern matching according to the fault feature vector, and determining the fault matching result; performing positioning verification on the low-voltage power distribution area according to the fault matching result, generating fault positioning parameters, so as to realize fast and efficient power outage fault diagnosis and positioning in the low-voltage power distribution area, and improve the positioning accuracy and reliability of the technical effect.

[0092] Embodiment 2, as Figure 2 is a schematic structural diagram of the power outage fault judgment and positioning system of the present invention for low-voltage power distribution areas. For example, Figure 1 In the present invention, the flow schematic diagram of the power outage fault judgment and positioning method for low-voltage power distribution areas can be implemented through a structure such as Figure 2 shown.

[0093] Based on the same concept as the power outage fault judgment and positioning method for low-voltage power distribution areas in the above embodiment, the present invention also provides a power outage fault judgment and positioning system for low-voltage power distribution areas, including:

[0094] An event data acquisition module 11, configured to collect power outage event data of a low-voltage power distribution area in real time.

[0095] A topological relationship construction module 12, configured to perform multi-dimensional feature analysis on the power outage event data, and construct a topological relationship network of the power distribution area according to spatio-temporal correlation features.

[0096] A feature vector generation and matching module 13, configured to generate a fault feature vector based on the spatio-temporal correlation features and the topological relationship network of the power distribution area, perform pattern matching according to the fault feature vector, and determine the fault matching result.

[0097] A positioning verification module 14, configured to perform positioning verification on the low-voltage power distribution area according to the fault matching result, and generate fault positioning parameters.

[0098] In some embodiments, the event data acquisition module 11 includes:

[0099] A voltage and current waveform data acquisition unit, configured to extract data from a low-voltage power distribution area through an intelligent meter terminal to obtain voltage and current waveform data.

[0100] A load mutation data receiving unit, configured to receive load mutation data of a distribution transformer monitoring terminal according to the voltage and current waveform data.

[0101] A power loss information determination unit, configured to retrieve an electricity consumption information data set, match the electricity consumption information data set with the load mutation data, and determine a plurality of power loss information.

[0102] A power outage event data generation unit for performing timestamp alignment processing on the multiple power loss information to determine the power outage event data.

[0103] In some embodiments, the topology relationship construction module 12 includes:

[0104] A geographic information parsing and device connection relationship map construction unit for introducing geographic information data of a low-voltage power distribution area for parsing, and constructing a device connection relationship map based on the geographic parsing data in combination with operation, distribution, and dispatching ledger information.

[0105] A spatial topology tree structure generation unit for generating a spatial topology tree structure based on the physical coordinates of the installation positions of smart meters and line branch boxes.

[0106] A power distribution area topology relationship network annotation unit for annotating the spatial topology tree structure according to the device connection relationship map to construct the power distribution area topology relationship network.

[0107] In some embodiments, the feature vector generation and matching module 13 includes:

[0108] A normalized eigenvalue generation unit for performing normalization processing on the spatio-temporal correlation features to generate normalized eigenvalues.

[0109] A multi-dimensional feature weight distribution parameter generation unit for performing weight calculation based on the power distribution area topology relationship network in combination with the normalized eigenvalues to generate multi-dimensional feature weight distribution parameters.

[0110] A fault feature vector generation unit for performing matrix splicing on the normalized eigenvalues and the multi-dimensional feature weight distribution parameters to generate the fault feature vector.

[0111] In some embodiments, the feature vector generation and matching module 13 further includes:

[0112] A fault mode library formulation unit for retrieving multiple historical fault case data, performing power outage fault analysis based on the multiple historical fault case data, and formulating a fault mode library.

[0113] A fault mode similarity coefficient acquisition unit for using the fault feature vector as an index to perform retrieval calculation on the fault mode library to obtain a fault mode similarity coefficient.

[0114] An associated fault instruction trigger unit for setting a fault threshold according to the multiple historical fault case data, and triggering an associated fault instruction when the fault mode similarity coefficient is greater than or equal to the fault threshold.

[0115] A fault matching result determination unit for backtracking the associated fault instruction to the fault mode library for mode matching to determine the fault matching result.

[0116] In some embodiments, the positioning verification module 14 includes:

[0117] A data conflict detection and grading identification unit, configured to perform data conflict detection on the low-voltage power distribution area based on the associated fault instruction, perform equipment grading identification on the low-voltage power distribution area according to the conflict detection result, generate a grading result, where the conflict detection result includes a candidate fault point set.

[0118] A fault point sequence generation unit, configured to perform weighted analysis on the grading result according to the fault matching result, obtain an equipment weight coefficient, perform probability sorting on the candidate fault point set according to the equipment weight coefficient, and generate a fault point sequence.

[0119] A fault location parameter determination unit, configured to perform verification based on the fault point sequence in combination with the grading result, and determine the fault location parameter.

[0120] In some embodiments, the positioning verification module 14 further includes:

[0121] A power outage impact range and repair priority determination unit, configured to perform impact analysis based on the fault point sequence, determine the power outage impact range, and formulate a repair priority according to the power outage impact range.

[0122] A fault repair path generation unit, configured to match the repair priority with the fault point sequence to generate a fault repair path.

[0123] A power outage fault repair strategy formulation unit, configured to traverse and check the duplicate of the fault location parameter according to the fault repair path, and formulate a power outage fault repair strategy.

[0124] It should be understood that the embodiments mentioned in this specification focus on their differences from other embodiments. The specific embodiments in the foregoing Embodiment 1 are equally applicable to the power outage fault research and judgment positioning system for low-voltage power distribution areas described in Embodiment 2. For the sake of brevity of the specification, no further elaboration is made here.

[0125] It should be understood that the disclosed embodiments of the present invention and the above descriptions can enable those skilled in the art to implement the present invention using the present invention. At the same time, the present invention is not limited to the part of the embodiments mentioned above. It should be understood that those of ordinary skill in the art can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A power outage fault judgment and location method for low-voltage distribution areas, characterized in that, The method includes: Collecting power outage event data of the low-voltage substation area in real time; Performing multi-dimensional feature analysis on the power outage event data, and constructing a substation area topology relationship network according to the spatio-temporal correlation features; Generating a fault feature vector based on the spatio-temporal correlation features and the substation area topology relationship network, performing pattern matching according to the fault feature vector, and determining a fault matching result; Performing positioning verification on the low-voltage substation area according to the fault matching result, and generating fault positioning parameters.

2. The power outage fault judgment and positioning method for low-voltage power distribution areas according to claim 1, wherein, Collecting power outage event data of the low-voltage substation area in real time, the method includes: Extracting data from the low-voltage substation area through an intelligent meter terminal to obtain voltage and current waveform data; Receiving load mutation data of a distribution transformer monitoring terminal according to the voltage and current waveform data; Retrieving an electricity consumption information data set, matching the electricity consumption information data set with the load mutation data, and determining multiple power loss information; Performing timestamp alignment processing on the multiple power loss information to determine the power outage event data.

3. The power outage fault judgment and positioning method for low-voltage power distribution areas according to claim 1, characterized in that Performing multi-dimensional feature analysis on the power outage event data, and constructing a substation area topology relationship network according to the spatio-temporal correlation features, the method includes: Introducing geographical information data of the low-voltage substation area for analysis, and constructing a device connection relationship graph according to the geographical analysis data combined with the operation, distribution and adjustment ledger information; Generating a spatial topology tree structure according to the physical coordinates of the intelligent meter installation position and the line branch box; Marking the spatial topology tree structure according to the device connection relationship graph to construct the substation area topology relationship network.

4. The power outage fault judgment and positioning method for low-voltage power distribution areas according to claim 1, characterized in that Generating a fault feature vector based on the spatio-temporal correlation features and the substation area topology relationship network, the method includes: Performing normalization processing on the spatio-temporal correlation features to generate a normalized feature value; Calculating weights based on the substation area topology relationship network combined with the normalized feature value to generate multi-dimensional feature weight distribution parameters; Performing matrix splicing on the normalized feature value and the multi-dimensional feature weight distribution parameters to generate the fault feature vector.

5. The power outage fault judgment and positioning method for low-voltage power distribution areas according to claim 1, characterized in that, Performing pattern matching according to the fault feature vector to determine a fault matching result, the method includes: Retrieving multiple historical fault case data, performing power outage fault analysis based on the multiple historical fault case data, and formulating a fault mode library; Using the fault feature vector as an index to perform retrieval calculation on the fault mode library to obtain a fault mode similarity coefficient; Setting a fault threshold according to the multiple historical fault case data, and triggering an associated fault instruction when the fault mode similarity coefficient is greater than or equal to the fault threshold; Retrospecting the associated fault instruction to the fault mode library for pattern matching to determine the fault matching result.

6. The power outage fault judgment and location method for low-voltage power distribution areas according to claim 5, characterized in that, Performing positioning verification on the low-voltage substation area according to the fault matching result to generate fault positioning parameters, the method includes: Performing data conflict detection on the low-voltage substation area based on the associated fault instruction, performing device grading identification on the low-voltage substation area according to the conflict detection result, and generating a grading result, where the conflict detection result includes a candidate fault point set; Performing weighted analysis on the grading result according to the fault matching result to obtain a device weight coefficient, and performing probability sorting on the candidate fault point set according to the device weight coefficient to generate a fault point sequence; Based on the fault point sequence and combined with the grading result, perform verification to determine the fault location parameters.

7. The power outage fault judgment and positioning method for low-voltage power distribution areas according to claim 6, characterized in that After generating the fault location parameters, the method includes: Perform impact analysis based on the fault point sequence to determine the power outage impact range, and formulate the repair priority according to the power outage impact range; Match the repair priority with the fault point sequence to generate a fault repair path; Traverse and check the fault location parameters according to the fault repair path to formulate a power outage fault repair strategy.

8. The power outage fault judgment and positioning system for low-voltage power distribution areas is characterized in that A system for executing the power outage fault judgment and location method for low-voltage power distribution areas according to any one of claims 1-7, the system includes: An event data acquisition module for real-time acquisition of power outage event data of the low-voltage power distribution area; A topology relationship construction module for performing multi-dimensional feature analysis on the power outage event data and constructing a power distribution area topology relationship network according to the spatio-temporal correlation features; A feature vector generation and matching module for generating a fault feature vector based on the spatio-temporal correlation features and the power distribution area topology relationship network, and performing pattern matching according to the fault feature vector to determine a fault matching result; A location verification module for performing location verification on the low-voltage power distribution area according to the fault matching result to generate fault location parameters.

Images