AI-based Substation Location Information Verification Method and System

Abstract

This invention discloses a method and system for verifying substation location information based on AI recognition and fusion, belonging to the field of artificial intelligence technology. The method includes: after hardware connection verification is completed in substation location information verification, selecting a first location for telemetry and remote control testing to obtain a first telemetry accuracy and a first remote control accuracy; selecting electrical connection points of the first location according to a moving window, performing iterative telemetry and remote control testing, and constructing a telemetry accuracy matrix and a remote control accuracy matrix; calling historical location verification data of points in the same family, and verifying the accuracy of the telemetry accuracy matrix and the remote control accuracy matrix to obtain a test reliability matrix; and performing AI fusion recognition based on the test reliability matrix, the telemetry accuracy matrix, and the remote control accuracy matrix to obtain a location verification quality parameter matrix, which serves as the substation location information verification result. This invention effectively improves the accuracy and efficiency of substation location information verification.

Description

Technical Field ,

[0008] , , ,

[0007] , ,

[0006] , ,

[0005] , ,

[0004] , , ,

[0010] ,

[0009]

[0001] The present invention relates to the technical field of artificial intelligence, and particularly to a method and system for verifying substation point information based on AI recognition fusion. Background Art

[0002] With the upgrade of the power system to intelligence and automation, the substation, as the core hub of power transmission, its operation stability directly determines the safety of power supply. After the substation is put into use, testing and verifying the remote control and telemetry functions of the internal electrical connection structure and information points becomes a key operation and maintenance link to ensure the operation quality. At present, the verification method in the industry is mainly direct testing, that is, by carrying out telemetry and remote control tests on a single point to collect parameters and complete the basic information verification work.

[0003] However, the existing verification technology only focuses on the direct test parameters of a single point, does not consider the relevance of electrical connections between points, nor does it verify the reliability of test data. The single-mode is easily affected by factors such as instantaneous fluctuations of equipment and environmental interference, resulting in accidental errors, and then leading to deviations in the verification results, unable to accurately reflect the true operating state of substation points, thus affecting the accuracy and efficiency of substation point information verification, and even may pose potential safety hazards for subsequent operation and maintenance. Summary of the Invention

[0004] The present invention provides a method and system for verifying substation point information based on AI recognition fusion, aiming to solve the technical problem of insufficient accuracy and efficiency in verifying substation point information in the prior art.

[0005] In view of the above problems, the present invention provides a method and system for verifying substation point information based on AI recognition fusion.

[0006] In the first aspect, the present invention provides a method for verifying substation point information based on AI recognition fusion, including:

[0007] After the hardware connection verification in the verification of substation point information is completed, select a first point for telemetry test and remote control test to obtain the first telemetry accuracy and the first remote control accuracy;

[0008] According to a moving window, select the points electrically connected to the first point for iterative telemetry test and remote control test, and construct a telemetry accuracy matrix and a remote control accuracy matrix;

[0009] Call the historical point verification data of the same-family points to verify the accuracy of the telemetry accuracy matrix and the remote control accuracy matrix, and obtain a test credibility matrix;

[0010] Based on the test credibility matrix, telemetry accuracy matrix, and remote control accuracy matrix, AI fusion recognition is performed to obtain the point verification quality parameter matrix, which serves as the substation location information verification result.

[0011] Secondly, this invention provides a substation location information verification system based on AI recognition and fusion, comprising:

[0012] The first point accuracy test module is used to select the first point for telemetry and remote control tests after the hardware connection verification is completed in the substation location information verification, and to obtain the first telemetry accuracy and the first remote control accuracy.

[0013] The iterative testing module is used to select the electrical connection point of the first point according to the moving window, perform iterative telemetry and remote control tests, and construct the telemetry accuracy matrix and the remote control accuracy matrix.

[0014] The accuracy verification module is used to call historical point verification data of the same family of points, perform accuracy verification on the telemetry accuracy matrix and the remote control accuracy matrix, and obtain the test confidence matrix.

[0015] The AI ​​fusion recognition module is used to perform AI fusion recognition based on the test confidence matrix, telemetry accuracy matrix and remote control accuracy matrix to obtain the point verification quality parameter matrix, which serves as the substation location information verification result.

[0016] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0017] This invention provides a method and system for verifying substation location information based on AI recognition and fusion. First, it conducts telemetry and remote control tests on a first location to obtain basic accuracy data. Then, it iteratively tests related locations using a moving window and constructs a double-precision matrix. Next, it calls historical data from locations within the same data family to verify the matrix's reliability. Finally, it uses AI to fuse multi-matrix data to generate a quality parameter matrix. This approach avoids the random errors of single tests and achieves comprehensive data correlation and accurate verification. This invention enhances the comprehensiveness and accuracy of location status assessment, improves the automation level and processing efficiency of verification, thereby effectively improving the accuracy and efficiency of substation location information verification. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1A flowchart illustrating the substation location information verification method based on AI recognition and fusion provided in an embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram of the structure of the substation location information verification system based on AI recognition and fusion provided in an embodiment of the present invention;

[0021] The components represented by each number in the attached diagram are explained below:

[0022] The system consists of a first-point accuracy testing module 11, an iterative testing module 12, an accuracy verification module 13, and an AI fusion recognition module 14. Detailed Implementation

[0023] This invention provides a method and system for verifying substation location information based on AI recognition and fusion, which addresses the technical problem of insufficient accuracy and efficiency in the verification of substation location information in existing technologies.

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0025] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0026] Example 1, as Figure 1 As shown, this invention provides a method for verifying substation location information based on AI recognition and fusion, the method comprising:

[0027] S100: After the hardware connection verification is completed in the substation location information verification, the first location is selected for telemetry and remote control testing to obtain the first telemetry accuracy and the first remote control accuracy.

[0028] In this embodiment of the invention, after the hardware connection verification in the substation location information verification is completed, a first location is selected for telemetry and remote control testing to obtain a first telemetry accuracy and a first remote control accuracy. The core of substation location information verification is to verify the matching between the correctness of the hardware connection and the accuracy of the information functions. The hardware connection is the physical basis for the information location to realize telemetry and remote control functions. If the hardware connection does not match the design drawings, subsequent telemetry and remote control tests will be meaningless, and may even cause equipment damage due to incorrect connection. Therefore, the hardware connection verification must be completed and confirmed to be correct before functional testing is carried out on the information location. The first location is selected as the initial test point to establish benchmark accuracy data, providing a reference for the iterative testing of subsequent related locations and avoiding test confusion caused by the lack of a benchmark.

[0029] Step S100 in the method provided in this embodiment of the invention includes:

[0030] After the hardware connection verification in the substation location information verification is completed and the verification result is correct, the substation design document is read, and the first point for telemetry and remote control is selected.

[0031] According to the test standards of the electrical components corresponding to the first point, telemetry and remote control tests are performed to obtain the first telemetry accuracy and the first remote control accuracy.

[0032] First, after verifying the hardware connections in the substation location information check and confirming the accuracy of the verification results, the substation design documents are read to select the first point for telemetry and remote control testing. Hardware connection verification refers to verifying the consistency between the physical connections of electrical components and the design documents, which is a prerequisite for information-based testing. Electrical components include cables, terminal blocks, and transformers. Physical connections include cable type, terminal number, and port correspondence. The first point is the reference point for initial testing, usually selected as the information-based point corresponding to the core functional components, facilitating subsequent correlation testing. Core functional components include transformers and circuit breakers. After completing the hardware connection verification of electrical components within the substation and confirming that the connection relationships are completely consistent with the design drawings and that there are no incorrect or missing connections, the substation design documents are read to select representative core information-based points as the first point for conducting telemetry and remote control testing. For example, verifying the secondary cable of the current transformer TA1 in the 10kV outgoing cabinet of a 110kV substation, confirming its connection to the AI1 terminal of the measurement and control device and the secondary wiring. Figure 1 Hardware verification passed. Read the point code table and determine that the information point code corresponding to TA1 is P101, whose function is to remotely monitor current and control protection setting on / off. Set P101 as the first point.

[0033] Secondly, in accordance with the test standards for the electrical components corresponding to the first point, telemetry and remote control tests are conducted to obtain the first telemetry accuracy and the first remote control accuracy.

[0034] Specifically, in accordance with the test standards for the electrical components corresponding to the first location, telemetry and remote control tests are performed to obtain the first telemetry accuracy and the first remote control accuracy, including:

[0035] According to the test standard of the electrical component corresponding to the first point, preset variables are set for the first point, telemetry test is performed to obtain the first telemetry variable, and the first point is remotely controlled by preset control commands to obtain the first remote control result.

[0036] Calculate the similarity between the first telemetry variable and the preset variable, and use it as the first telemetry accuracy;

[0037] Calculate the similarity between the first remote control result and the preset control result within the preset control command to obtain the first remote control accuracy.

[0038] First, according to the testing standards for the electrical components corresponding to the first point, preset variables are set for the first point, and telemetry testing is conducted to obtain the first telemetry variable. Simultaneously, preset control commands are sent to the first point to obtain the first remote control result. The electrical component testing standards refer to documents that clearly define the range of test parameters and operating procedures. Preset variables refer to the baseline parameter values ​​for telemetry testing, such as current and voltage, used to determine the accuracy of data acquisition. Preset control commands are standard control commands for the function of the point. The first telemetry variable refers to the parameter value actually acquired by the measurement and control system. The first remote control result refers to the actual state of the component after responding to the command. Based on the industry testing standards for the electrical components corresponding to the first point, preset variables that conform to the component's operating range are set for this point. Telemetry testing is conducted through the measurement and control system, and the actually acquired first telemetry variable is recorded. Simultaneously, preset control commands are sent to this point, the component's response state is observed, and the first remote control result is recorded.

[0039] For example, according to the electrical component testing standard, a preset variable of 400A simulated current is injected into point P101. After telemetry testing, the actual current value is 398A, that is, the first telemetry variable is 398A. A preset control command is sent to P101: complete the activation of the TA1 overcurrent protection setting within 10 seconds, and the monitoring and control system displays "Activation successful"; it is actually observed that the protection device is activated in 12 seconds, and the monitoring and control system fully displays "Activation successful". The first remote control result is recorded as: 12-second response, status display "Activation successful".

[0040] Secondly, the similarity between the first telemetry variable and the preset variable is calculated as the first telemetry accuracy. Telemetry accuracy is an indicator that quantifies the accuracy of telemetry function, with a value ranging from 0-100%. The higher the value, the more accurate the telemetry. The similarity between the first telemetry variable and the preset variable is calculated using a quantitative formula, and the similarity is directly used as the first telemetry accuracy. The formula is: Telemetry Accuracy = (1 - |First Telemetry Variable - Preset Variable| / Preset Variable) × 100%. For example, if the preset variable is 400A and the first telemetry variable is 398A, substituting into the formula: Telemetry Accuracy = (1 - |398 - 400| / 400) × 100% = (1 - 2 / 400) × 100% = 99.5%, that is, the first telemetry accuracy is 99.5%. This transforms the deviation of the telemetry data into an intuitive accuracy indicator, avoiding ambiguous judgments and providing a quantitative conclusion for the accuracy of the telemetry function of the first point.

[0041] Finally, the similarity between the first remote control result and the preset control result within the preset control command is calculated to obtain the first remote control accuracy. Remote control accuracy is an indicator that quantifies the accuracy of the remote control function response, ranging from 0-100%, reflecting the degree of fit between the command response and the preset result. First, the key evaluation dimensions of the preset control result are broken down, and a weight is assigned to each dimension. Then, the fit between the first remote control result and the preset result in each dimension is compared, and a weighted average score is calculated according to the weights. This average score is the remote control result similarity and is directly used as the first remote control accuracy. The specific formula is: First remote control accuracy = Σ(fit score of a certain dimension × weight of that dimension) × 100%. Wherein, the fit score is quantified as follows: completely consistent = 1 point, partially consistent = 0.5 points, completely inconsistent = 0 points. For example, the preset control command is: complete the activation of the TA1 overcurrent protection setting within 10 seconds, and the measurement and control system displays "activated successfully"; the first remote control result is: a 12-second response, with the status displaying "activated successfully". Evaluation dimensions and weights: execution status weight 0.6; response time weight 0.4. Compatibility assessment: Execution status fully compliant, 1 point; response time 12 seconds exceeds the standard, 0.5 points. Calculation result: First remote control accuracy = 1 × 0.6 + 0.5 × 0.4 × 100% = (0.6 + 0.2) × 100% = 80%.

[0042] In this embodiment of the invention, by first confirming the correctness of the hardware foundation and then selecting benchmark points for testing, the problem of invalid testing caused by incorrect hardware connection is eliminated from the source. At the same time, through clear standards and quantified precision calculations, accurate initial benchmark data is provided for subsequent iterative testing of related points, avoiding precision deviations caused by testing without reference, and initially ensuring the basic accuracy of substation location information verification.

[0043] S200: According to the moving window, select the electrical connection point of the first point, perform iterative telemetry test and remote control test, and construct the telemetry accuracy matrix and the remote control accuracy matrix.

[0044] In this embodiment of the invention, points electrically connected to the first point are selected according to the moving window, and iterative telemetry and remote control tests are performed to construct telemetry accuracy matrices and remote control accuracy matrices. The information points within a substation are not isolated but are closely interconnected through electrical circuits. An abnormal telemetry / remote control accuracy at a certain point may be caused by a fault in a point directly connected to it. Testing only a single first point cannot reflect the quality of the entire point network. Therefore, it is necessary to focus on direct electrical connections, iteratively test all related points using a moving window, and simultaneously integrate the scattered accuracy data into a matrix based on location coordinates. This avoids the one-sidedness of isolated testing and provides structured data support for subsequent accuracy verification based on topology association.

[0045] Step S200 in the method provided in this embodiment of the invention includes:

[0046] According to the moving window of direct electrical connection, select the point of electrical connection of the first point as multiple second points;

[0047] Telemetry and remote control tests were performed on multiple second points to obtain multiple second telemetry accuracies and multiple second remote control accuracies.

[0048] Continue testing to obtain the telemetry accuracy and remote control accuracy of all points within the substation. Based on the location coordinates of all points, construct the telemetry accuracy matrix and the remote control accuracy matrix.

[0049] First, following the moving window for direct electrical connections, select points electrically connected to the first test point as multiple second test points. The moving window for direct electrical connections refers to a dynamic filtering range centered on the current test point, containing only points directly connected by conductors, ensuring a physical electrical connection between the test object and the initial point. Second test points are the associated points directly electrically connected to the first test point, serving as the initial extension objects for iterative testing. Using the first test point as the initial center, define the moving window for direct electrical connections, filtering only points directly connected to the current test point without intermediate electrical components, and selecting all information-based points within this range as second test points. After completing the testing within this window, repeat the window filtering logic using each second test point as a new center until all points within the substation are covered.

[0050] For example, taking the first point P101 as the initial center, the direct electrical connection points are screened according to the secondary wiring diagram: the remote control point P102 of the circuit breaker QF1 directly receives the current signal of the current transformer TA1; the telemetry point P103 of the disconnector switch QS1 shares the same power supply circuit with TA1; the telemetry point P104 corresponding to the analog input terminal AI2 of the measurement and control device directly collects the secondary side signal of TA1; the above P102, P103, and P104 are the first batch of second points. After their testing is completed, P102 is used as the new center to screen the directly connected P105, and so on iteratively. The direct electrical connection window accurately locks the associated points, avoiding irrelevant points from interfering with the test efficiency; the iterative logic ensures that all points are covered without omission, while preserving the topological relationship between points.

[0051] Secondly, telemetry and remote control tests are performed on multiple second points to obtain multiple second telemetry accuracies and multiple second remote control accuracies. Second telemetry / remote control accuracy: This is a quantitative indicator of the telemetry / remote control accuracy of the second points. The calculation logic is completely consistent with that of the first point, ensuring data comparability. Following the test standards and accuracy calculation logic of step S100, for each second point, industry test standards are first consulted based on the model of its corresponding electrical component. Adaptive preset variables and preset control commands are set, and after the test is executed, the second telemetry / remote control results are recorded. Finally, the telemetry and remote control accuracy of each second point are calculated using the same formula.

[0052] For example, taking the second point P102 (QF1 circuit breaker, function: remote tripping / telemetry tripping current) as an example, the telemetry test is as follows: According to the electrical component test standard, a preset variable tripping current of 10A is injected into P102, and the actual second telemetry variable is 9.8A. The second telemetry accuracy is calculated as (1-|9.8-10| / 10)×100%=98%; The remote control test is as follows: A preset control command is sent: QF1 tripping is completed within 15 seconds, and the control system displays "Successful tripping". The actual result is: tripping in 13 seconds, displaying "tripping". The second remote control accuracy is calculated as (0.5×0.6+1×0.4)×100%=70%; Similarly, the tests for P103 and P104 are completed, and the results are recorded as follows: P103 (second telemetry accuracy 99%, second remote control accuracy 90%), P104 (second telemetry accuracy 99.3%, second remote control accuracy none, marked "telemetry only"). By adopting unified testing and calculation standards, we ensure that the accuracy data of all points are consistent, providing a reliable data foundation for subsequent matrix integration and cross-point analysis.

[0053] Finally, the telemetry and remote control accuracy of all points within the substation were obtained through further testing. Based on the location coordinates of all points, a telemetry accuracy matrix and a remote control accuracy matrix were constructed. The telemetry accuracy matrix is ​​a two-dimensional structured data set indexed by location coordinates. It stores the telemetry accuracy of all points according to the index relationship of cabinet number + floor number - point code, intuitively reflecting the distribution of telemetry quality at different locations. The remote control accuracy matrix has the same structure as the telemetry accuracy matrix, but only stores the remote control accuracy data of all points. Points without remote control functions are marked with "-" for distinction. The moving window was expanded from each measured point as the center, gradually covering all 120 information-based points within the substation, completing the calculation of telemetry and remote control accuracy for all points. Extract the 3D coordinates of each location (cabinet number, floor number, terminal number) and construct a 2D matrix structure. Use the cabinet number + floor number as the row index, such as C10-L1, C10-L2, C10-L3, corresponding to floors 1-3 of cabinet C10. Use the location code as the column index, such as P101-P115, corresponding to the 15 locations in cabinet C10. Fill the corresponding row and column intersection of the telemetry accuracy matrix with the telemetry accuracy value of each location, and fill the corresponding position of the remote control accuracy matrix with the remote control accuracy value of each location. For locations with only telemetry function and no remote control function, mark "-" in the corresponding position of the remote control accuracy matrix to clarify the functional attribute.

[0054] For example, after completing the testing of all 120 points, focus on cabinet C10, which contains 15 points. Cabinet number C10, floor numbers L1-L3, and point codes P101-P115 are used. The row indices of its telemetry accuracy matrix are set to C10-L1, C10-L2, and C10-L3, and the column indices are set to P101 to P115. Data is entered according to the corresponding indices: the intersection of row C10-L2 and column P101 is filled with the telemetry accuracy of P101 (99.5%); the intersection of row C10-L2 and column P102 is filled with the telemetry accuracy of P102 (98%); the intersection of row C10-L1 and column P103 is filled with the telemetry accuracy of P101... The telemetry accuracy of 03 is 99%, and the intersection of row C10-L3 and column P104 is filled with the telemetry accuracy of P104, which is 99.3%. The remote control accuracy matrix is ​​filled with data according to the same index logic: 80% is filled at the intersection of row C10-L2 and column P101, 70% is filled at the intersection of row C10-L2 and column P102, 90% is filled at the intersection of row C10-L1 and column P103, and "-" is marked at the intersection of row C10-L3 and column P104. The matrix data of cabinets C11 to C19 is filled in sequentially according to the above logic, and finally the telemetry accuracy matrix and remote control accuracy matrix covering all points of the entire substation are formed.

[0055] In this embodiment of the invention, the iterative testing logic of the moving window via direct electrical connection ensures coverage of all points while preserving the electrical topology relationships, avoiding potential omissions caused by isolated testing; the unified standard testing process guarantees the consistency of accuracy data; and the construction of a double-precision matrix based on location coordinates transforms scattered data into structured assets, realizing both visualized management of point quality and providing core data support for subsequent topology matching and accuracy verification of points within the same family, effectively improving the efficiency and accuracy of subsequent analysis.

[0056] S300: Call the historical point verification data of the same family of points, perform accuracy verification on the telemetry accuracy matrix and the remote control accuracy matrix, and obtain the test reliability matrix.

[0057] In this embodiment of the invention, historical data from points within the same network are used to verify the accuracy of the telemetry accuracy matrix and the remote control accuracy matrix, thereby obtaining a test reliability matrix. Points directly electrically connected within a substation exhibit strong correlation; a change in the accuracy of one point is transmitted to related points through electrical circuits, forming a characteristic pattern of accuracy change. Existing tests only acquire the accuracy of a single point, without verifying whether its variation with related points conforms to conventional patterns, potentially leading to data distortion due to accidental interference. Therefore, it is necessary to calculate the accuracy variation of related points, combine it with historical data from points within the same network topology for verification, determine the rationality of the current accuracy data, and ultimately form a test reliability matrix, providing a reliable data quality assessment basis for subsequent AI fusion recognition.

[0058] Step S300 in the method provided in this embodiment of the invention includes:

[0059] Based on the telemetry accuracy matrix and the remote control accuracy matrix, calculate the variation range of telemetry accuracy and remote control accuracy of the first point and multiple second points to obtain multiple first associated telemetry variation ranges and multiple first associated remote control variation ranges;

[0060] Construct a connection topology between the first point and multiple second points. Extract family points with the same connection topology from the historical point verification data of other substations. Verify and analyze the multiple first associated telemetry change amplitudes and multiple first associated remote control change amplitudes to obtain the first test credibility.

[0061] Continue to verify the telemetry and remote control accuracy of other locations, obtain the test reliability of all locations, and construct a test reliability matrix.

[0062] First, based on the telemetry accuracy matrix and remote control accuracy matrix, the variation ranges of telemetry accuracy and remote control accuracy for the first point and multiple second points are calculated to obtain multiple first associated telemetry variation ranges and multiple first associated remote control variation ranges. The first associated telemetry variation range refers to the absolute difference between the telemetry accuracy of the second point and the first point, reflecting the degree of fluctuation in telemetry accuracy between associated points. The first associated remote control variation range is the absolute difference between the remote control accuracy of the second point and the first point, reflecting the degree of fluctuation in remote control accuracy between associated points. Using the accuracy data of the first point as a benchmark, the difference in telemetry accuracy and the difference in remote control accuracy between each second point and the first point are calculated respectively. The absolute value of this difference is the associated accuracy variation range, corresponding to the first associated telemetry variation range and the first associated remote control variation range, respectively; for points without remote control functionality, only the telemetry variation range is calculated.

[0063] For example, given that the telemetry accuracy of the first point P101 is 99.5% and the remote control accuracy is 80%, the accuracy data of the second point is as follows: P102 (telemetry accuracy 98%, remote control accuracy 70%): First associated telemetry change amplitude = |98%-99.5%| = 1.5%; First associated remote control change amplitude = |70%-80%| = 10%; P103 (telemetry accuracy 99%, remote control accuracy 90%): First associated telemetry change amplitude = |99%-99.5%| = 0.5%; First associated remote control change amplitude = |90%-80%| = 10%; P104 (telemetry accuracy 99.3%, no remote control): First associated telemetry change amplitude = |99.3%-99.5%| = 0.2%; The first associated remote control change amplitude is marked with "-"; Finally, the set of first associated telemetry change amplitudes is obtained as {1.5%, 0.5%, 0.2%}, and the set of first associated remote control change amplitudes is obtained as {10%, 10%, -}. By quantifying the accuracy variation of related points, the implicit electrical correlation between points is transformed into intuitive data, providing core analytical indicators for subsequent comparison with the patterns of points in the same family.

[0064] Secondly, a connection topology is constructed between the first location and multiple second locations. Within the historical location verification data of other substations, family-like locations with the same connection topology are extracted. The change amplitudes of the multiple first associated telemetry and the multiple first associated remote control are verified and analyzed to obtain the first test credibility.

[0065] Specifically, a connection topology is constructed between a first location and multiple second locations. Within the historical location verification data of other substations, family-like locations with the same connection topology are extracted. Verification analysis is performed on the multiple first-related telemetry change amplitudes and multiple first-related remote control change amplitudes to obtain the first test reliability, including:

[0066] Construct a connection topology between the first location and multiple second locations, and extract family-like locations with the same connection topology from the historical location verification data of other substations;

[0067] The average values ​​of the associated telemetry change amplitude and the associated remote control change amplitude of the points in the same family are obtained during telemetry and remote control tests, and are used as the associated telemetry change amplitude and the associated remote control change amplitude of multiple points in the same family.

[0068] The similarity between the multiple first associated telemetry change amplitudes and the multiple family-related associated telemetry change amplitudes is calculated to obtain multiple telemetry test confidence levels.

[0069] Calculate the similarity between the variation amplitudes of the multiple first associated remote controls and the variation amplitudes of the multiple related remote controls to obtain the reliability of multiple remote control tests;

[0070] Based on multiple telemetry test confidence levels and multiple remote control test confidence levels, the first test confidence level of the first location is calculated.

[0071] First, a connection topology is constructed for the first substation location and multiple second substation locations. From historical data of other substation locations, family-like locations with the same connection topology are extracted. The connection topology refers to a structural diagram that visually represents the direct electrical connection path between the first substation location and the second substation locations, clarifying the relationships between the locations. Family-like locations refer to a set of locations in the historical database that are completely identical to the current location group in terms of voltage level, component combination, and connection topology; their correlation and change patterns are relevant. Based on the direct electrical connection relationship between the first substation location and multiple second substation locations, the connection path between the first and second substation locations is graphically defined, forming the connection topology. Subsequently, the historical database is searched to select other substation locations with the same voltage level, component combination, and connection topology structure as family-like locations.

[0072] For example, construct the connection topology: Using P101 (TA1) as the core, draw a topology diagram: P101 (TA1) directly connects to P102 (QF1), P101 (TA1) directly connects to P103 (QS1), and P101 (TA1) directly connects to P104 (AI2), clearly defining three sets of direct connections. Extract related points: From historical databases, such as data from 50 110kV substations, select 12 groups of points that meet the following requirements: 10kV voltage level, current transformer (TA) + circuit breaker (QF) + disconnector (QS) + monitoring and control device component combination, and a connection topology consistent with the above. These are designated as related points for P101. The connection topology locks in the core association of the current point group. The selection of related points ensures the comparability of subsequent comparative data, providing a reliable reference benchmark for pattern verification.

[0073] Secondly, the average values ​​of the associated telemetry and remote control variation amplitudes of points within the same family were obtained during telemetry and remote control testing. These were used as the multiple associated telemetry and remote control variation amplitudes for the same family. The associated telemetry variation amplitude refers to the average historical associated telemetry variation amplitude of corresponding connections within a group of points within the same family. The associated remote control variation amplitude refers to the average historical associated remote control variation amplitude of corresponding connections within a group of points within the same family. For each corresponding connection relationship within a group of points within the same family, historical telemetry / remote control variation amplitude data was extracted, and the average value for each corresponding relationship was calculated, serving as the associated telemetry variation amplitude and the associated remote control variation amplitude for the same family, respectively.

[0074] For example, for three sets of corresponding connections within the same data set, the historical correlation variation amplitude for each set is extracted: TA-QF: historical telemetry variation amplitude data {1.1%, 1.3%, 1.2%...}, with a calculated mean of 1.2%; historical remote control variation amplitude data {7.8%, 8.2%, 8.0%...}, with a mean of 8%; TA-QS: historical telemetry variation amplitude mean of 0.6%, historical remote control variation amplitude mean of 12%; TA-AI: historical telemetry variation amplitude mean of 0.3%, with "-" marked for no remote control function; finally, the set of related telemetry variation amplitudes within the same data set is obtained as {1.2%, 0.6%, 0.3%}, and the set of related remote control variation amplitudes within the same data set is {8%, 12%, -}. By calculating the mean, the common variation patterns of data sets within the same data set are extracted, providing a quantitative benchmark for judging the rationality of the current variation amplitude and avoiding the random interference of a single set of historical data.

[0075] Furthermore, the similarity between the multiple first-related telemetry change amplitudes and the multiple family-related telemetry change amplitudes is calculated to obtain multiple telemetry test reliability scores. Telemetry test reliability is the degree of fit between the telemetry change amplitude of a single group of related points and the family mean, ranging from 0-100%. A higher similarity indicates that the current telemetry change better conforms to the electrical correlation pattern, and the lower the possibility of random error. Similarity = 1 - |current change amplitude - family mean| / family mean × 100%. The similarity between the first-related telemetry change amplitude of each group and the corresponding family-related telemetry change amplitude is calculated, and each similarity score represents the telemetry test reliability score for the corresponding connection relationship.

[0076] For example, given the known first-related telemetry variation amplitudes {1.5% (P101-P102), 0.5% (P101-P103), 0.2% (P101-P104)}, and the family mean {1.2%, 0.6%, 0.3%}: P101-P102: Similarity = 1 - |1.5% - 1.2%| / 1.2% × 100% ≈ 97.5%, i.e., telemetry test reliability 97.5%), P1 P1-P103: Similarity = 1 - |0.5% - 0.6%| / 0.6% × 100% ≈ 98.3%, i.e., telemetry test reliability is 98.3%; P101-P104: Similarity = 1 - |0.2% - 0.3%| / 0.3% × 100% ≈ 96.7%, i.e., telemetry test reliability is 96.7%; finally, multiple sets of telemetry test reliability are obtained {97.5%, 98.3%, 96.7%}. By quantifying similarity, the regularity of telemetry variation amplitude is verified, intuitively distinguishing between cases conforming to electrical correlation and possible random errors, providing a detailed basis for the overall reliability.

[0077] Simultaneously, the similarity between the multiple first-related remote control change amplitudes and the multiple family-related remote control change amplitudes is calculated to obtain multiple remote control test credibility scores. Remote control test credibility refers to the degree of fit between the remote control change amplitude of a single group of related points and the family average, with a value ranging from 0-100%, reflecting the rationality of the current remote control change. Following the calculation logic of telemetry test credibility, the same similarity formula is used to calculate the degree of fit between the first-related remote control change amplitude and the corresponding family-related remote control change amplitude for each group, obtaining the corresponding remote control test credibility score for each group; points without remote control functionality are marked with "-" and are not included in the calculation.

[0078] For example, given the known first associated remote control variation amplitudes {10% (P101-P102), 10% (P101-P103), -}, and the average variation amplitudes of related remote controls {8%, 12%, -}: P101-P102: similarity = 1 - |10%-8%| / 8%×100% = 97.5%, i.e., remote control test reliability 97.5%; P101-P103: similarity = 1 - |10%-12%| / 12%×100%≈98.3%, i.e., remote control test reliability 98.3%; finally, multiple remote control test reliability sets {97.5%, 98.3%} are obtained. This complements the telemetry test reliability, comprehensively covering the telemetry and remote control functions of associated points, ensuring that the rationality of both types of accuracy data is verified.

[0079] Finally, based on the reliability of multiple telemetry tests and multiple remote control tests, the first test reliability of the first location is calculated. The first test reliability refers to the final index quantifying the reliability of the accuracy data of the first location, taking a value of 0-100%, which integrates the verification results of multiple sets of related telemetry and remote control tests. The weights of both telemetry and remote control test reliability are set to 0.5. First, the average values ​​of multiple telemetry test reliability and multiple remote control test reliability are calculated separately, and then they are weighted and summed to obtain the first test reliability of the first location.

[0080] For example, the average reliability of telemetry is calculated as: (97.5% + 98.3% + 96.7%) / 3 ≈ 97.5%; the average reliability of remote control is calculated as: (97.5% + 98.3%) / 2 ≈ 97.9%; and the weighted reliability of the first test is calculated as: (97.5% × 0.5 + 97.9% × 0.5) × 100% ≈ 97.7%. By averaging and weighting multiple sets of subdivided reliability, the correlation between telemetry and remote control is comprehensively considered, avoiding the influence of single-set data bias, and making the reliability of the first test more objective and convincing.

[0081] Finally, the telemetry and remote control accuracy of other points are verified to obtain the test reliability of all points, and a test reliability matrix is ​​constructed. The test reliability matrix is ​​a structured data set that is completely consistent with the telemetry / remote control accuracy matrix and stores the reliability indicators of all points after verification by the same family topology. Following the verification logic of the first point, each point in the substation that is not the core point is taken as the new core point. The complete process is executed in sequence: screening directly electrically connected related points, calculating the telemetry / remote control correlation change amplitude between the core and related points, constructing the connection topology of the core point, extracting the historical average change amplitude of the same family points, calculating multiple sets of telemetry / remote control test reliability, and weighting to obtain the test reliability of the core point. After the test reliability of all points is calculated, the test reliability of each point is filled into the corresponding position according to the unified index structure of the telemetry / remote control accuracy matrix in S200. The reliability of points without remote control function is calculated separately according to the telemetry test reliability, and finally a test reliability matrix covering the entire substation is formed.

[0082] For example, iteratively verify the new core point P102 (QF1): Associated points: P101 (TA1), P105 (QF1 auxiliary contact point); Accuracy data: P102 (telemetry accuracy 98%, remote control accuracy 70%), P101 (telemetry accuracy 99.5%, remote control accuracy 80%), P105 (telemetry accuracy 99.2%, remote control accuracy 85%); Associated change range: P102-P101 (associated telemetry change range 1.5%, associated remote control change range 10%), P102-P105 (associated telemetry change range 1.2%, associated remote control change range 15%); Connection topology: P102—P101, P102—P105; Average remote control change range of the same family: P102-P101 (telemetry 1.3%, remote control 9%), P102-P105 ... 0.0%, remote control 14%); Test confidence: average confidence of telemetry 98.25%, average confidence of remote control 99.1%, final confidence ≈ 98.7%; Similarly, iterate on other points: P103 (QS1): associated with P101 and P106, test confidence 97.9%; P104 (AI2): telemetry only, associated with P101, test confidence 98.3%; P105: associated with P102, test confidence 97.5%; Construct a matrix (C10 cabinet as an example): C10-L2 row: P101 (97.7%), P102 (98.7%), P105 (97.5%); C10-L1 row: P103 (97.9%); C10-L3 row: P104 (98.3%); Complete the filling of points in cabinets C11-C19, forming a test confidence matrix covering 120 points.

[0083] In this embodiment of the invention, by calculating the magnitude of correlation changes, topologically matching family points, and verifying similarity, isolated point precision data is transformed into reliable data after correlation pattern verification, effectively filtering out distorted data caused by accidental interference; the test reliability matrix and the preceding double precision matrix have the same structure, realizing a one-to-one correspondence between data quality and precision data, providing a key basis for weight allocation in subsequent AI fusion recognition, and further improving the accuracy of substation location information verification results.

[0084] S400: Based on the test confidence matrix, telemetry accuracy matrix and remote control accuracy matrix, perform AI fusion recognition to obtain the point verification quality parameter matrix, which serves as the substation location information verification result.

[0085] In this embodiment of the invention, AI fusion recognition is performed based on the test credibility matrix, telemetry accuracy matrix, and remote control accuracy matrix to obtain a point verification quality parameter matrix, which serves as the substation location information verification result. Telemetry accuracy and remote control accuracy are two core dimensions reflecting the quality of substation locations. The former reflects the accuracy of parameter acquisition, while the latter reflects the reliability of command response. Single-dimensional data cannot comprehensively characterize the actual operational quality of the location. AI fusion recognition can deeply mine the inherent correlation characteristics of the two types of accuracy data, outputting a more comprehensive basic quality assessment. The test credibility matrix has been verified through family topology and can accurately reflect the reliability of accuracy data. Using the test credibility matrix to correct the basic assessment results can filter out distorted data caused by random errors, making the final point verification quality parameters more realistic and further improving the accuracy and reliability of the verification results.

[0086] Step S400 in the method provided in this embodiment of the invention includes:

[0087] Based on the telemetry accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain the basic point verification quality parameter matrix;

[0088] Based on the test credibility matrix, the basic point verification quality parameter matrix is ​​corrected to obtain the point verification quality parameter matrix, which serves as the verification result of substation location information.

[0089] First, based on the telemetry accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain the basic point verification quality parameter matrix.

[0090] Specifically, based on the telemetry accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain a basic point verification quality parameter matrix, including:

[0091] Based on the historical substation location information verification records, sample telemetry accuracy sets and sample remote control accuracy sets are collected, and the actual location electrical quality parameters after verification are collected and labeled to obtain the sample location verification quality parameter set.

[0092] A point verification and analysis network was constructed based on AI machine learning.

[0093] Using the sample telemetry accuracy set, sample remote control accuracy set, and sample location verification quality parameter set, the location verification analysis network is supervised and trained and tested until the test accuracy is qualified.

[0094] The telemetry accuracy and remote control accuracy of the same points within the telemetry accuracy matrix and remote control accuracy matrix are combined and input into the point verification analysis network to output the basic point verification quality parameters, thereby obtaining the basic point verification quality parameter matrix.

[0095] First, based on historical substation location information verification records, sample telemetry accuracy sets and sample remote control accuracy sets are collected. Simultaneously, verified electrical quality parameters for the actual locations are collected and labeled to obtain a sample location verification quality parameter set. The sample telemetry accuracy set refers to the historical telemetry accuracy data set for each location, used by the AI ​​model to learn the correlation between telemetry accuracy and actual quality. The sample remote control accuracy set refers to the historical remote control accuracy data set for each location, used by the AI ​​model to learn the correlation between remote control accuracy and actual quality. The sample location verification quality parameter set refers to the sample set labeled with actual quality parameters, used for supervised training of the AI ​​model. From the location information verification records of multiple substations of the same voltage level over a historical period, sample telemetry accuracy and sample remote control accuracy for all locations are collected to form sample telemetry accuracy sets and sample remote control accuracy sets. Simultaneously, the actual electrical quality parameters for these locations, verified through manual review or equipment operation, are extracted as sample location verification quality parameters and labeled to form the sample location verification quality parameter set.

[0096] For example, sample data from 12,000 points were collected from the historical records of 50 110kV substations: The sample telemetry accuracy set includes data such as 99.2% historical telemetry accuracy for P101 in cabinet C10 and 97.8% historical telemetry accuracy for P102; the sample remote control accuracy set includes data such as 82% historical remote control accuracy for P101 in cabinet C10 and 71% historical remote control accuracy for P102; and the sample point verification quality parameter set includes historical true quality parameters such as 95 points for P101 and 90 points for P102, with higher scores indicating better point quality. This provides sufficient training samples for the AI ​​model, ensuring that the model can learn the inherent correlation between telemetry, remote control accuracy, and the true quality of the points.

[0097] Secondly, a location verification analysis network is constructed based on AI machine learning. This network is a specialized AI model designed to integrate telemetry and remote control accuracy features and output location quality parameters. Using AI machine learning technology, a neural network structure suitable for numerical feature fusion, such as a multilayer perceptron, is selected to construct the network. The input layer consists of a two-dimensional feature vector representing the combined telemetry and remote control accuracy of a single location. The hidden layer extracts the fused features through an activation function. The output layer outputs a single numerical location verification quality parameter, ranging from 0 to 100 points, with higher scores indicating better quality. This builds an AI model architecture adapted to dual-precision fusion, laying the foundation for accurate quality parameter output. For example, a three-layer neural network is constructed: an input layer with two neurons receiving telemetry and remote control accuracy data respectively; a hidden layer with 16 neurons using the ReLU activation function; and an output layer with one neuron, outputting a quality parameter ranging from 0 to 100 points. The core function of the network is to learn the intrinsic correlation between telemetry, remote control accuracy, and the actual location quality.

[0098] Furthermore, the aforementioned sample telemetry accuracy set, sample remote control accuracy set, and sample point verification quality parameter set are used to supervise the training and testing of the point verification analysis network until the test accuracy is satisfactory. Supervised training refers to training the model with sample data bearing real labels, allowing the model to learn the explicit correlation between input and output. The test accuracy reflects the degree of fit between the model's predicted quality parameters and the actual quality parameters; the higher the accuracy, the better the model performance. The sample telemetry accuracy set and sample remote control accuracy set are combined in a single-point two-dimensional feature vector format as model input; the sample point verification quality parameter set is used as the model output label for supervised training. Network parameters are adjusted through backpropagation to reduce the error between the predicted value and the actual label. The model accuracy is verified using a test set every 10 training rounds until the test accuracy is ≥95%, at which point training is stopped and the model is saved.

[0099] For example, 6000 samples were used as the training set and 1500 samples as the test set. Initially, the model's accuracy was 72%. After 50 rounds of training, the accuracy stabilized at 96.3%, meeting the ≥95% passing standard. Training was then stopped, and the trained point verification analysis network was saved. By optimizing model parameters through training, the model was ensured to accurately uncover the correlation between double-precision data and point quality, providing a guarantee for outputting reliable basic quality parameters.

[0100] Subsequently, the telemetry accuracy and remote control accuracy of the unified points within the telemetry accuracy matrix and remote control accuracy matrix are combined and input into the point verification analysis network. This outputs basic point verification quality parameters, resulting in a basic point verification quality parameter matrix. The basic point verification quality parameters refer to the preliminary point quality assessment values ​​output by the AI ​​model after integrating telemetry and remote control accuracy, ranging from 0 to 100 points. The basic point verification quality parameter matrix is ​​a structured data set storing the basic quality parameters of all points according to a unified index, consistent with the structure of the preceding matrix. From the telemetry accuracy matrix and remote control accuracy matrix, the telemetry accuracy and remote control accuracy of each point are extracted and combined into a two-dimensional feature vector for the unified point. The feature vectors of all points are sequentially input into the trained point verification analysis network, outputting the basic point verification quality parameters for each point. The basic quality parameters are filled into the corresponding positions according to the unified index structure of cabinet number + floor number - point code, constructing the basic point verification quality parameter matrix. By integrating dual-precision data through AI models, scattered precision indicators are transformed into unified quality parameters, enabling a comprehensive preliminary assessment of site quality and providing basic data for subsequent corrections.

[0101] For example, extract the two-dimensional feature vectors of the C10 cabinet locations: P101: telemetry accuracy 99.5%, remote control accuracy 80%; P102: telemetry accuracy 98%, remote control accuracy 70%; P103: telemetry accuracy 99%, remote control accuracy 90%; P104: telemetry accuracy 99.3%, remote control function missing; P105: telemetry accuracy 99.2%, remote control accuracy 85%. Input these into the trained location verification and analysis network to output basic quality parameters: P101: 94 points, P102: 88 points, P103: 95 points, P104: 92 points, P105: 93 points. Construct a basic matrix by index. Taking the C10 cabinet as an example: C10-L2 row: P101 (94 points), P102 (88 points), P105 (93 points); C10-L1 row: P103 (95 points); C10-L3 row: P104 (92 points).

[0102] Secondly, based on the test credibility matrix, the basic point verification quality parameter matrix is ​​corrected to obtain the point verification quality parameter matrix, which serves as the substation location information verification result.

[0103] Specifically, based on the test reliability matrix, the basic point verification quality parameter matrix is ​​corrected to obtain the point verification quality parameter matrix, which serves as the substation location information verification result, including:

[0104] Based on the test confidence matrix, configure the point verification quality correction coefficient matrix;

[0105] The point verification quality correction coefficient matrix is ​​used to correct the basic point verification quality parameter matrix to obtain the point verification quality parameter matrix, which serves as the substation location information verification result.

[0106] First, based on the test credibility matrix, configure the point-to-point verification quality correction coefficient matrix. The point-to-point verification quality correction coefficient refers to the adjustment coefficient set according to the test credibility, used to correct basic quality parameters and reflect the impact of the reliability of accuracy data on quality assessment. The point-to-point verification quality correction coefficient matrix has the same structure as the preceding matrix, storing a structured data set of all point-to-point correction coefficients. Based on the test credibility of each point in the test credibility matrix, set the corresponding point-to-point verification quality correction coefficient. The higher the test credibility, the closer the correction coefficient is to 1, indicating that the basic quality parameters are more reliable and no significant adjustment is needed; the lower the test credibility, the further the correction coefficient deviates from 1, indicating that the basic quality parameters may be distorted and targeted adjustments are required. For example, set the following rules: credibility ≥ 98%, correction coefficient 1.0; 95% ≤ credibility < 98%, correction coefficient 0.99; 90% ≤ credibility < 95%, correction coefficient 0.98; credibility < 90%, correction coefficient 0.95. Fill the correction coefficients into the corresponding positions according to cabinet number + floor number - point code index to construct the point-to-point verification quality correction coefficient matrix. Alternatively, the test reliability can be directly used as the quality correction coefficient for point verification. For example, if the reliability is 98%, the correction coefficient is 0.98. Converting test reliability into a quantifiable correction coefficient provides a clear basis for the precise adjustment of basic quality parameters and ensures that the correction logic is traceable.

[0107] For example, the test confidence of the C10 cabinet location is extracted as follows: P101: 97.7%, P102: 98.7%, P103: 97.9%, P104: 98.3%, P105: 97.5%; the correction coefficients are configured according to the rules as follows: P101 (97.7%): 0.99, P102 (98.7%): 1.0, P103 (97.9%): 0.99, P104 (98.3%): 1.0, P105 (97.5%): 0.99; the correction coefficient matrix is ​​constructed (C10 cabinet as an example): C10-L2 row: P101 (0.99), P102 (1.0), P105 (0.99); C10-L1 row: P103 (0.99); C10-L3 row: P104 (1.0).

[0108] Secondly, the aforementioned point-to-point verification quality correction coefficient matrix is ​​used to correct and calculate the basic point-to-point verification quality parameter matrix, obtaining the point-to-point verification quality parameter matrix as the substation location information verification result. The final point-to-point verification quality parameter refers to the point quality assessment value after reliability correction through testing, which more closely reflects the actual operational quality of the point. The point-to-point verification quality parameter matrix is ​​a structured data set storing the final quality parameters of all points and is the core result of this verification. The final point-to-point verification quality parameter = basic point-to-point verification quality parameter × point-to-point verification quality correction coefficient. Correction calculations are performed on the data of each point in the basic point-to-point verification quality parameter matrix; according to the unified index structure of cabinet number + floor number - point code, the corrected final quality parameters are filled into the corresponding positions to form the point-to-point verification quality parameter matrix, which serves as the final result of the substation location information verification.

[0109] For example, the final quality parameters of cabinet C10 are corrected and calculated as follows: P101: 94 points × 0.99 ≈ 93 points; P102: 88 points × 1.0 = 88 points; P103: 95 points × 0.99 ≈ 94 points; P104: 92 points × 1.0 = 92 points; P105: 93 points × 0.99 ≈ 92 points; Construct the final matrix, taking cabinet C10 as an example: C10-L2 row: P101 (93 points), P102 (88 points), P105 (92 points); C10-L1 row: P103 (94 points); C10-L3 row: P104 (92 points); Complete the correction calculation and matrix filling for all points of cabinets C11-C19 in sequence, and finally form a point verification quality parameter matrix covering 120 points of the entire substation. By correcting the coefficients to filter out the distortion of precision data, the evaluation bias caused by the distortion is made so that the final quality parameters are closer to the actual state of the location, thus ensuring the reliability of the verification results.

[0110] In this embodiment of the invention, AI-based fusion recognition integrates telemetry and remote control data to overcome the limitations of single-dimensional assessment and achieve a comprehensive preliminary evaluation of site quality. Corrections are made based on a test reliability matrix, effectively filtering out distorted data caused by random errors, ensuring that the final site verification quality parameter matrix is ​​both comprehensive and accurate. The site verification quality parameter matrix has a unified structure with the preceding matrix, intuitively presenting the true quality level of each site in the entire substation, providing maintenance personnel with clear verification results, and ultimately effectively improving the accuracy and efficiency of substation site information verification.

[0111] Through the specific implementation methods described above, the embodiments of the present invention achieve the following technical effects:

[0112] This invention provides a method and system for verifying substation location information based on AI-based recognition and fusion. First, a reliable origin is established through single-point benchmark testing. Second, iterative testing of electrically related point groups is performed using a moving window to construct a precision matrix covering the network, improving testing efficiency. Next, historical data from the same family is introduced for cross-validation to generate a reliability matrix, effectively suppressing random errors and enhancing data reliability. Finally, AI fusion technology is used to intelligently analyze and weight the multi-source matrix, outputting a comprehensive and quantitative quality parameter matrix. This invention realizes a transformation from isolated single-point testing to networked evaluation, and from manual judgment to intelligent decision-making. While improving the automation and efficiency of testing, it enhances the accuracy and precision of evaluation results through multi-dimensional verification and fusion analysis.

[0113] Example 2, as Figure 2 As shown, this invention provides a substation location information verification system based on AI recognition and fusion, the system comprising:

[0114] The first point accuracy test module 11 is used to select the first point for telemetry and remote control tests after the hardware connection verification is completed in the substation location information verification, so as to obtain the first telemetry accuracy and the first remote control accuracy.

[0115] The iterative test module 12 is used to select the electrical connection point of the first point according to the moving window, perform iterative telemetry test and remote control test, and construct the telemetry accuracy matrix and the remote control accuracy matrix;

[0116] The accuracy verification module 13 is used to call the historical point verification data of the same family of points, perform accuracy verification on the telemetry accuracy matrix and the remote control accuracy matrix, and obtain the test confidence matrix.

[0117] The AI ​​fusion recognition module 14 is used to perform AI fusion recognition based on the test credibility matrix, telemetry accuracy matrix and remote control accuracy matrix to obtain the point verification quality parameter matrix, which serves as the substation location information verification result.

[0118] In one embodiment, the first point accuracy testing module 11 is further configured to:

[0119] After the hardware connection verification in the substation location information verification is completed and the verification result is correct, the substation design document is read, and the first point for telemetry and remote control is selected.

[0120] According to the test standards of the electrical components corresponding to the first point, telemetry and remote control tests are performed to obtain the first telemetry accuracy and the first remote control accuracy.

[0121] Specifically, in accordance with the test standards for the electrical components corresponding to the first location, telemetry and remote control tests are performed to obtain the first telemetry accuracy and the first remote control accuracy, including:

[0122] According to the test standard of the electrical component corresponding to the first point, preset variables are set for the first point, telemetry test is performed to obtain the first telemetry variable, and the first point is remotely controlled by preset control commands to obtain the first remote control result.

[0123] Calculate the similarity between the first telemetry variable and the preset variable, and use it as the first telemetry accuracy;

[0124] Calculate the similarity between the first remote control result and the preset control result within the preset control command to obtain the first remote control accuracy.

[0125] In one embodiment, the iterative testing module 12 is further configured to:

[0126] According to the moving window of direct electrical connection, select the point of electrical connection of the first point as multiple second points;

[0127] Telemetry and remote control tests were performed on multiple second points to obtain multiple second telemetry accuracies and multiple second remote control accuracies.

[0128] Continue testing to obtain the telemetry accuracy and remote control accuracy of all points within the substation. Based on the location coordinates of all points, construct the telemetry accuracy matrix and the remote control accuracy matrix.

[0129] In one embodiment, the precision verification module 13 is further configured to:

[0130] Based on the telemetry accuracy matrix and the remote control accuracy matrix, calculate the variation range of telemetry accuracy and remote control accuracy of the first point and multiple second points to obtain multiple first associated telemetry variation ranges and multiple first associated remote control variation ranges;

[0131] Construct a connection topology between the first point and multiple second points. Extract family points with the same connection topology from the historical point verification data of other substations. Verify and analyze the multiple first associated telemetry change amplitudes and multiple first associated remote control change amplitudes to obtain the first test credibility.

[0132] Continue to verify the telemetry and remote control accuracy of other locations, obtain the test reliability of all locations, and construct a test reliability matrix.

[0133] Specifically, a connection topology is constructed between a first location and multiple second locations. Within the historical location verification data of other substations, family-like locations with the same connection topology are extracted. Verification analysis is performed on the multiple first-related telemetry change amplitudes and multiple first-related remote control change amplitudes to obtain the first test reliability, including:

[0134] Construct a connection topology between the first location and multiple second locations, and extract family-like locations with the same connection topology from the historical location verification data of other substations;

[0135] The average values ​​of the associated telemetry change amplitude and the associated remote control change amplitude of the points in the same family are obtained during telemetry and remote control tests, and are used as the associated telemetry change amplitude and the associated remote control change amplitude of multiple points in the same family.

[0136] The similarity between the multiple first associated telemetry change amplitudes and the multiple family-related associated telemetry change amplitudes is calculated to obtain multiple telemetry test confidence levels.

[0137] Calculate the similarity between the variation amplitudes of the multiple first associated remote controls and the variation amplitudes of the multiple related remote controls to obtain the reliability of multiple remote control tests;

[0138] Based on multiple telemetry test confidence levels and multiple remote control test confidence levels, the first test confidence level of the first location is calculated.

[0139] In one embodiment, the AI ​​fusion recognition module 14 is further used for:

[0140] Based on the telemetry accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain the basic point verification quality parameter matrix;

[0141] Based on the test credibility matrix, the basic point verification quality parameter matrix is ​​corrected to obtain the point verification quality parameter matrix, which serves as the verification result of substation location information.

[0142] Specifically, based on the telemetry accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain a basic point verification quality parameter matrix, including:

[0143] Based on the historical substation location information verification records, sample telemetry accuracy sets and sample remote control accuracy sets are collected, and the actual location electrical quality parameters after verification are collected and labeled to obtain the sample location verification quality parameter set.

[0144] A point verification and analysis network was constructed based on AI machine learning.

[0145] Using the sample telemetry accuracy set, sample remote control accuracy set, and sample location verification quality parameter set, the location verification analysis network is supervised and trained and tested until the test accuracy is qualified.

[0146] The telemetry accuracy and remote control accuracy of the same points within the telemetry accuracy matrix and remote control accuracy matrix are combined and input into the point verification analysis network to output the basic point verification quality parameters, thereby obtaining the basic point verification quality parameter matrix.

[0147] Specifically, based on the test reliability matrix, the basic point verification quality parameter matrix is ​​corrected to obtain the point verification quality parameter matrix, which serves as the substation location information verification result, including:

[0148] Based on the test confidence matrix, configure the point verification quality correction coefficient matrix;

[0149] The point verification quality correction coefficient matrix is ​​used to correct the basic point verification quality parameter matrix to obtain the point verification quality parameter matrix, which serves as the substation location information verification result.

[0150] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0151] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0152] This specification and accompanying drawings are merely illustrative examples of the invention and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its scope. Therefore, if such modifications and modifications fall within the scope of the invention and its equivalents, the invention is intended to include these modifications and modifications.

Claims

1. A substation site information checking method based on AI recognition fusion, characterized in that, The method comprises: After the hardware connection check in the substation point information check is completed, a first point is selected for remote measurement test and remote control test to obtain a first remote measurement accuracy and a first remote control accuracy; According to a moving window, a point connected to the first point is selected for iterative remote measurement test and remote control test, and a remote measurement accuracy matrix and a remote control accuracy matrix are constructed; The historical point check data of the same family point is called to perform accuracy verification on the remote measurement accuracy matrix and the remote control accuracy matrix to obtain a test credibility matrix; According to the test credibility matrix, the remote measurement accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain a point check quality parameter matrix as a substation point information check result; The historical point check data of the same family point is called to perform accuracy verification on the remote measurement accuracy matrix and the remote control accuracy matrix to obtain a test credibility matrix, comprising: According to the remote measurement accuracy matrix and the remote control accuracy matrix, the change amplitudes of the remote measurement accuracy and the remote control accuracy of the first point and a plurality of second points are calculated to obtain a plurality of first associated remote measurement change amplitudes and a plurality of first associated remote control change amplitudes; A connection topology of the first point and the plurality of second points is constructed, and the same family points having the same connection topology are extracted from the historical point check data of other substations, and the plurality of first associated remote measurement change amplitudes and the plurality of first associated remote control change amplitudes are verified and analyzed to obtain a first test credibility; The accuracy verification on the remote measurement accuracy and the remote control accuracy of other points is continued to obtain the test credibility of all points, and a test credibility matrix is constructed; According to the test credibility matrix, the remote measurement accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain a point check quality parameter matrix as a substation point information check result, comprising: According to the remote measurement accuracy matrix and the remote control accuracy matrix, AI fusion recognition is performed to obtain a basic point check quality parameter matrix; According to the test credibility matrix, the basic point check quality parameter matrix is corrected to obtain a point check quality parameter matrix as a substation point information check result. 2.The AI recognition fusion-based substation site information checking method according to claim 1, characterized in that, After the hardware connection check in the substation point information check is completed, a first point is selected for remote measurement test and remote control test to obtain a first remote measurement accuracy and a first remote control accuracy, comprising: After the hardware connection check in the substation point information check is completed and the check result is correct, a substation design file is read, and a first point for remote measurement and remote control is selected; According to the test standard of the electrical element corresponding to the first point, remote measurement test and remote control test are performed to obtain a first remote measurement accuracy and a first remote control accuracy. 3.The AI recognition fusion-based substation site information checking method according to claim 2, characterized in that, According to the test standard of the electrical element corresponding to the first point, remote measurement test and remote control test are performed to obtain a first remote measurement accuracy and a first remote control accuracy, comprising: According to the test standard of the electrical element corresponding to the first point, a preset variable is set for the first point, remote measurement test is performed to obtain a first remote measurement variable, and remote control of the first point is performed with a preset control instruction to obtain a first remote control result; The similarity of the first remote measurement variable and the preset variable is calculated as the first remote measurement accuracy; Calculate the similarity between the first remote control result and the preset control result in the preset control instruction to obtain the first remote control accuracy. 4.The AI recognition fusion-based substation site information checking method according to claim 1, characterized in that, According to the moving window, select the point connected to the first point for iterative remote measurement test and remote control test, and construct a remote measurement accuracy matrix and a remote control accuracy matrix, including: According to the moving window of direct electrical connection, select the points electrically connected to the first point as a plurality of second points; Perform remote measurement test and remote control test on the plurality of second points to obtain a plurality of second remote measurement accuracies and a plurality of second remote control accuracies; Continue testing to obtain the remote measurement accuracy and the remote control accuracy of all points in the substation, and construct a remote measurement accuracy matrix and a remote control accuracy matrix based on the position coordinates of all points. 5.The AI recognition fusion-based substation site information checking method according to claim 1, characterized in that, Construct the connection topology of the first point and the plurality of second points, extract the same family points with the same connection topology from the historical point checking data of other substations, and verify and analyze the plurality of first associated remote measurement change amplitudes and the plurality of first associated remote control change amplitudes to obtain the first test credibility, including: Construct the connection topology of the first point and the plurality of second points, extract the same family points with the same connection topology from the historical point checking data of other substations; Obtain the average associated remote measurement change amplitude and the average associated remote control change amplitude of the same family points in the remote measurement and remote control test with the connected points as a plurality of same family associated remote measurement change amplitudes and a plurality of same family associated remote control change amplitudes; Calculate the similarity between the plurality of first associated remote measurement change amplitudes and the plurality of same family associated remote measurement change amplitudes to obtain a plurality of remote measurement test credibilities; Calculate the similarity between the plurality of first associated remote control change amplitudes and the plurality of same family associated remote control change amplitudes to obtain a plurality of remote control test credibilities; Based on the plurality of remote measurement test credibilities and the plurality of remote control test credibilities, calculate the first test credibility of the first point. 6.The AI recognition fusion-based substation site information checking method according to claim 1, characterized in that, According to the remote measurement accuracy matrix and the remote control accuracy matrix, perform AI fusion recognition to obtain a basic point checking quality parameter matrix, including: According to the substation point information checking record in the historical time, collect a sample remote measurement accuracy set and a sample remote control accuracy set, and collect the real point electrical quality parameters after checking and verification to label a sample point checking quality parameter set; Based on AI machine learning, construct a point checking analysis network; Use the sample remote measurement accuracy set, the sample remote control accuracy set, and the sample point checking quality parameter set to supervise the training and testing of the point checking analysis network until the test accuracy is qualified; Combine the remote measurement accuracy and the remote control accuracy of the same point in the remote measurement accuracy matrix and the remote control accuracy matrix, input the point checking analysis network, and output the basic point checking quality parameter to obtain a basic point checking quality parameter matrix. 7.The AI recognition fusion-based substation site information checking method according to claim 1, characterized in that, According to the test credibility matrix, correct the basic point checking quality parameter matrix to obtain a point checking quality parameter matrix as the substation point information checking result, including: According to the test credibility matrix, configure a point checking quality correction coefficient matrix; The point position collation quality parameter matrix is corrected by using the point position collation quality correction coefficient matrix, and a point position collation quality parameter matrix is obtained as a substation point position information collation result.

8. A substation site information checking system based on AI recognition fusion, characterized in that, The system for implementing the AI recognition fusion-based substation point position information collation method of any one of claims 1-7 comprises: The first point position precision test module is configured to, after the hardware connection collation in the substation point position information collation is completed, select a first point position for remote measurement test and remote control test, and obtain a first remote measurement precision and a first remote control precision. The iterative test module is configured to select point positions electrically connected to the first point position according to a moving window, perform iterative remote measurement test and remote control test, and construct a remote measurement precision matrix and a remote control precision matrix. The precision verification module is configured to call historical point position collation data of point positions of the same family, perform precision verification on the remote measurement precision matrix and the remote control precision matrix, and obtain a test credibility matrix. The AI fusion recognition module is configured to perform AI fusion recognition according to the test credibility matrix, the remote measurement precision matrix, and the remote control precision matrix, obtain a point position collation quality parameter matrix, and use the point position collation quality parameter matrix as a substation point position information collation result.

Images