Intelligent Topology-Based Dynamic Anti-Misoperation Method and System for Grid Equipment Startup

Abstract

The present invention discloses a method and system for dynamic anti-misoperation of power grid equipment startup based on intelligent topology, which relates to the technical field of dynamic anti-misoperation of power grid equipment startup, and includes the following steps: constructing a topology matrix of power grid equipment and fusing it with equipment parameters to obtain an anti-misoperation feature vector; performing time-domain analysis on the anti-misoperation feature vector to identify a lightning strike distortion feature vector; based on DTW path screening, complementing and correcting the lightning strike distortion feature vector; constructing an anti-misoperation decision model based on the complemented and corrected feature vector, and real-time outputting an anti-misoperation operation instruction, which solves the problem that the feature vector generated by the traditional anti-misoperation model directly based on distorted data will mis-represent the equipment state, resulting in false locking or false operation.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic anti-misoperation for grid equipment startup, and more specifically, to a dynamic anti-misoperation method and system for grid equipment startup based on intelligent topology. Background Art

[0002] With the increasing complexity and intelligence of the power system, problems such as transient fluctuations and state switching of grid equipment often occur during the startup and operation of grid equipment. Therefore, how to accurately identify faults and avoid misoperations in real time during the startup, switching, and operation of grid equipment has become an important challenge in power system protection.

[0003] In recent years, the dynamic anti-misoperation method for grid equipment based on intelligent topology has gradually become a research hotspot. By introducing real-time monitoring, data analysis, and intelligent algorithms, it is possible to analyze the electrical characteristics (such as impedance, phase difference, etc.) of equipment in real time during the startup process of grid equipment, construct a dynamic operation model of the equipment in combination with the topology map, and accurately judge the operation state of the equipment. In particular, the application of adaptive algorithms and transfer learning enables the anti-misoperation system to automatically optimize parameters according to the historical data of the equipment and the operation conditions of the power grid, improving the fault detection accuracy and response speed. The intelligent anti-misoperation mechanism can not only reduce false alarms and missed alarms, but also dynamically adjust strategies to ensure the high efficiency and safety of the power grid in the face of complex operating environments.

[0004] For example, an intelligent dispatching order anti-misoperation system disclosed in the invention patent with the publication number CN115800510A includes a topology relationship module, a section module, a verification module, and a server; wherein the topology relationship module is used to update the grid topology connection relationship in real time; the section module is used to analyze the grid topology connection relationship diagram according to the work content of the corresponding dispatcher to obtain a dispatching verification diagram; the verification module is used to verify the dispatching order. The specific method includes: establishing an order response library, obtaining the dispatching order received by the corresponding dispatcher, and matching the corresponding verification data from the order response library according to the obtained dispatching order; setting a simulation unit, and the dispatcher conducts dispatching simulation through the simulation unit to obtain a simulated operation, verifies the simulated operation with the obtained verification data to obtain a verification result, and gives corresponding operation prompts according to the obtained verification result.

[0005] For example, a method, system, device, and medium for preventing errors in power grid dispatching operations announced in the invention patent announcement with the announcement number CN113991843B constructs a knowledge graph of power grid dispatching operations, mines the correlation between the power grid operation graph and the device graph, and further refines the anti-error rules for dispatching operations through rule learning. Through an intelligent learning method and using rule learning technology, rule learning is carried out on a large number of maintenance graphs, operation graphs, and device graph cases in the graph; the learned rule library is used to check whether the newly opened operation ticket conforms, achieving the purpose of preventing errors, greatly reducing the manpower input for verification, and improving the operation efficiency.

[0006] In the above disclosed technical solution, there are at least the following technical problems:

[0007] Under strong electromagnetic interference such as lightning strikes, the original data of power grid equipment collected is prone to instantaneous distortion and become distorted. The traditional anti-error model directly generates feature vectors based on the distorted data, which will misrepresent the device state and thus lead to false locking or false operation. For example, the current spike caused by a lightning strike may be misjudged as a short-circuit fault, triggering an incorrect circuit breaker opening instruction. In response to the above problems, the present invention proposes a solution. Summary of the Invention

[0008] To overcome the above defects of the prior art, the embodiments of the present invention provide a method and system for dynamically preventing errors in the startup of power grid equipment based on intelligent topology. By constructing a power grid topology matrix and fusing device parameters, modeling is carried out after time-domain analysis and DTW data filling, and anti-error operation instructions are output in real time to solve the problem that the traditional anti-error model directly generates feature vectors based on distorted data, which will misrepresent the device state and thus lead to false locking or false operation.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] The method for dynamically preventing errors in the startup of power grid equipment based on intelligent topology includes the following steps: constructing a topology matrix of power grid equipment and fusing it with device parameters to obtain an anti-error feature vector; performing time-domain analysis on the anti-error feature vector to identify lightning strike distortion feature vectors; based on DTW path screening, filling in the lightning strike distortion feature vectors; constructing an anti-error determination model based on the filled-in feature vectors and outputting anti-error operation instructions in real time.

[0011] In a preferred embodiment, the constructing a topology matrix of power grid equipment and fusing it with device parameters to obtain an anti-error feature vector is specifically: obtaining real-time device parameter data of the target area and obtaining a topology matrix based on frequency-domain coupling degree analysis; performing feature fusion on the real-time device parameter data and the topology matrix to obtain an anti-error feature vector.

[0012] In a preferred embodiment, the time-domain analysis of the anti-error feature vector to identify the lightning strike distortion feature vector is specifically as follows: perform wavelet packet decomposition on the anti-error feature vector to extract the high-frequency band energy ratio feature; obtain the reference energy eigenvalue of the frequency band, and compare and analyze the ratio of the high-frequency band energy ratio feature to the reference energy eigenvalue of the frequency band with a first threshold to obtain the lightning strike distortion feature vector.

[0013] In a preferred embodiment, the filling-in of the lightning strike distortion feature vector based on DTW path screening is specifically as follows: perform time window truncation on the lightning strike distortion feature vector to obtain the truncated lightning strike distortion feature vector sequence; construct the standard trajectory sequence of the lightning strike feature, and use the DTW algorithm to perform path offset matching between the truncated lightning strike distortion feature vector sequence and the standard trajectory sequence of the lightning strike feature to obtain a number of first alignment paths; screen the number of first alignment paths to obtain the optimal alignment path; fill in the lightning strike distortion feature vector based on the optimal alignment path.

[0014] In a preferred embodiment, the screening of the number of first alignment paths to obtain the optimal alignment path is specifically as follows: calculate the local slope of each node in the number of first alignment paths with respect to the predecessor node; calculate the cumulative path bending energy of each first alignment path based on the preset cumulative path bending energy calculation formula; eliminate the first alignment paths whose cumulative path bending energy does not meet the preset constraint conditions to obtain the second alignment paths; perform secondary screening on the second alignment paths based on the time sequence constraint conditions and topological logic constraint conditions to obtain the optimal alignment path.

[0015] In a preferred embodiment, the filling-in of the lightning strike distortion feature vector based on the optimal alignment path is specifically as follows: analyze the optimal alignment path to obtain the time index of the distorted signal and the time index of the standard signal; map the time points of the distorted signal to the time points of the standard signal and synchronize them; correct the amplitude of the distorted signal and replace it with the corresponding amplitude of the standard signal to obtain the filled-in feature vector.

[0016] The technical effects and advantages of the method and system for starting dynamic anti-error of power grid equipment based on intelligent topology according to the present invention:

[0017] 1. By fusing real-time device parameter data with the topology matrix, the anti-error feature vector of the present invention is more accurate, can reflect the dynamic state of power grid equipment in real time, thereby improving the accuracy and real-time performance of misjudgment recognition. Secondly, using wavelet packet decomposition for time-domain analysis can extract the high-frequency band energy ratio feature, and through comparative analysis with the reference energy eigenvalue of the frequency band, the lightning strike distortion feature vector is successfully identified, increasing the sensitivity to short-term abnormal phenomena such as lightning strikes and improving the anti-error ability.

[0018] During the process of filling in missing values, based on the method of DTW path screening, by matching the path offset between the lightning strike distortion feature vector and the standard trajectory, the distortion features can be effectively corrected. Further path screening calculates the local slope and path bending energy, ensuring the accuracy of the filling-in-missing-values process. And under the constraints of considering time sequence and topological logic, the repair effect of the feature vector is further optimized, realizing fast and accurate anti-misoperation control under lightning strike interference, significantly reducing the risk of misoperation, and being applicable to high electromagnetic interference scenarios such as transmission lines and substations. Description of the Drawings

[0019] Figure 1 It is a schematic flow chart of the method for starting dynamic anti-misoperation of power grid equipment based on intelligent topology of the present invention.

[0020] Figure 2 It is a schematic structural diagram of the system for starting dynamic anti-misoperation of power grid equipment based on intelligent topology of the present invention.

[0021] Figure 3 It is a DTW signal alignment path broken line diagram regarding DTW path screening in the embodiment provided by the present invention. Detailed Embodiment

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0023] Embodiment 1 Figure 1 A method for dynamic anti-misoperation of power grid equipment based on intelligent topology of the present invention is given, including the following steps:

[0024] S1. Construct a topological matrix of power grid equipment and fuse it with equipment parameters to obtain an anti-misoperation feature vector;

[0025] In this example, constructing a topological matrix of power grid equipment and fusing it with equipment parameters to obtain an anti-misoperation feature vector is specifically as follows:

[0026] Obtain real-time equipment parameter data of the target area and obtain a topological matrix based on frequency domain coupling degree analysis;

[0027] Perform feature fusion on the real-time equipment parameter data and the topological matrix to obtain an anti-misoperation feature vector.

[0028] In this example, obtaining a topological matrix based on frequency domain coupling degree analysis is specifically as follows:

[0029] Perform Fourier transform on the current signal of the equipment to obtain the cross-spectrum density of the equipment;

[0030] Extract the spectral characteristic values of lightning strike interference through wavelet transform;

[0031] Calculate the frequency-domain coupling degree of the device under lightning strike interference based on the cross-spectral density of the device and the spectral characteristic values of lightning strike interference.

[0032] Among them, the specific formula for frequency-domain coupling degree analysis is as follows:

[0033]

[0034] Among them, is the frequency-domain coupling degree between device i and device j, is the frequency, is the lightning strike sensitive frequency band, is the cross-spectral density between device i and device j, is the spectral characteristic value of lightning strike interference, is a constant, is the cross-spectral density of device i, is the cross-spectral density of device j.

[0035] It should be noted that constructing a topology matrix by using frequency-domain coupling degree analysis can accurately reflect the mutual relationship of devices in different frequency bands. By performing Fourier transform on the current signals of the devices, the cross-spectral density of the devices can be obtained, which can provide a quantitative data basis for the coupling relationship between the devices. And wavelet transform extracts the spectral characteristic values of lightning strike interference, further enhancing the recognition ability of the method for external lightning strike interference. Combining these two pieces of information, the calculated frequency-domain coupling degree of the device can effectively identify the impact of lightning strike interference on the device behavior, improving the recognition sensitivity and accuracy of the system for interference signals; by performing feature fusion on the real-time device parameters and the topology matrix, the anti-misoperation feature vector can not only reflect the state of the power grid devices, but also take into account the impact of abnormal interferences such as lightning strikes on the devices. This way of feature fusion makes the recognition of misjudgments more comprehensive and detailed, not relying solely on a single parameter or signal, thus improving the anti-misoperation effect and reliability.

[0036] S2. Perform time-domain analysis on the anti-misoperation feature vector to identify the lightning strike distortion feature vector;

[0037] In this example, performing time-domain analysis on the anti-misoperation feature vector to identify the lightning strike distortion feature vector is specifically as follows:

[0038] Perform wavelet packet decomposition on the anti-misoperation feature vector to extract the high-frequency band energy ratio feature;

[0039] Obtain the reference energy characteristic value of the frequency band, compare and analyze the ratio of the high-frequency band energy ratio feature to the reference energy characteristic value of the frequency band with the first threshold to obtain the lightning strike distortion feature vector.

[0040] It should be noted that, first of all, time-domain analysis of the anti-error feature vector by wavelet packet decomposition can effectively extract the high-frequency band energy ratio feature of the signal. Wavelet packet decomposition is a multi-resolution analysis method that can perform signal processing simultaneously in the time domain and frequency domain, and is particularly suitable for the analysis of non-stationary signals. When the power grid system faces lightning interference, the signal usually shows distortion of high-frequency components. Through wavelet packet decomposition, these high-frequency features can be accurately identified, and the high-frequency band energy ratio can be extracted, thereby reflecting the distortion degree of the power grid equipment after being affected by lightning strikes.

[0041] Secondly, obtaining the band reference energy eigenvalue and comparing it with the high-frequency band energy ratio feature is the key step in identifying the lightning strike distortion feature vector. Through the ratio analysis with the band reference energy eigenvalue, the specific impact of lightning interference on the signal can be judged more accurately. This ratio can be compared with a preset first threshold. If the ratio exceeds the threshold, it indicates that the signal may be affected by lightning strike distortion, thereby effectively identifying the signal characteristics of lightning interference. This analysis method based on energy ratio is simple and efficient, can quickly locate the lightning strike distortion signal, and reduce the probability of misjudgment and missed judgment.

[0042] S3. Based on DTW path screening, fill in the missing parts of the lightning strike distortion feature vector;

[0043] In this example, based on DTW path screening, filling in the missing parts of the lightning strike distortion feature vector is specifically as follows:

[0044] Perform time window truncation on the lightning strike distortion feature vector to obtain the truncated lightning strike distortion feature vector sequence;

[0045] Construct the standard trajectory sequence of lightning strike features, and use the DTW algorithm to perform path offset matching between the truncated lightning strike distortion feature vector sequence and the standard trajectory sequence of lightning strike features to obtain a number of first alignment paths;

[0046] Screen the number of first alignment paths to obtain the optimal alignment path;

[0047] Fill in the missing parts of the lightning strike distortion feature vector based on the optimal alignment path.

[0048] It should be noted that performing time window truncation on the lightning strike distortion feature vector can focus the distorted part of the signal within a specific time range, reduce redundant information, and thus improve the accuracy of subsequent processing. The truncated lightning strike distortion feature vector sequence is the core data for processing, providing a relatively concentrated and targeted signal segment for subsequent path matching.

[0049] Next, by constructing a standard trajectory sequence of lightning strike characteristics and using the DTW algorithm for path offset matching, the non-linear alignment problem between time series can be solved. The DTW algorithm has the advantage of processing asynchronous time series and can flexibly match the lightning strike distortion feature vector sequence with the standard trajectory sequence to obtain multiple first alignment paths. These paths can display different matching schemes, thus providing multiple possibilities for selecting the most suitable alignment path.

[0050] After obtaining multiple alignment paths, through further screening, the optimal alignment path is finally selected. The screening process considers the overall matching quality of the paths and selects the best path through some quantitative criteria (such as the shape, distance, error, etc. of the path). This link ensures the accuracy of the filling process and avoids incorrect repairs caused by inappropriate paths.

[0051] Finally, based on the optimal alignment path, the lightning strike distortion feature vector is filled. Through the aligned path, the time points and amplitudes of the distorted signal can be accurately corrected to make it more consistent with the standard trajectory and restore the true form of the signal. This filling step ensures the accuracy of the filled feature vector, thereby improving the credibility of subsequent analysis and determination.

[0052] In this example, several first alignment paths are screened to obtain the optimal alignment path, specifically:

[0053] Calculate the local slope of each node and its predecessor node in several first alignment paths;

[0054] Based on the preset cumulative path bending energy calculation formula, calculate the cumulative path bending energy of each first alignment path;

[0055] Eliminate the first alignment paths whose cumulative path bending energy does not meet the preset constraint conditions to obtain the second alignment paths;

[0056] Perform a secondary screening on the second alignment paths based on the time sequence constraint conditions and topological logic constraint conditions to obtain the optimal alignment path.

[0057] It should be noted that calculating the local slope of each node and its predecessor node in each first alignment path helps to evaluate the smoothness and stability of the path. The local slope reflects the rate of change of the path. A large or irregular local slope may indicate poor alignment quality of the path. Through this index, those paths that deviate far from the standard trajectory can be initially screened out, potential incorrect alignments can be eliminated, and thus the accuracy of the subsequent filling process can be improved.

[0058] Next, based on the preset cumulative path bending energy calculation formula, the overall quality of each first alignment path is evaluated. The cumulative path bending energy quantifies the twists and deviations generated during the alignment process. An alignment path with a lower path bending energy is usually smoother and has a higher degree of coincidence with the standard trajectory. Through this index, paths with excessive curvature and large deviations can be further filtered out to ensure that the paths used in the interpolation process can accurately reflect the true form of the lightning strike distorted signal.

[0059] Then, the paths that do not meet the preset cumulative path bending energy constraint conditions are removed to obtain the second alignment paths. This step further improves the accuracy of path screening by removing paths that do not meet the quality standards, ensuring that the finally used paths are smoother and meet the expected alignment standards.

[0060] Finally, a secondary screening is performed based on the chronological order and topological logic constraint conditions, further refining the selection of the optimal path.

[0061] Among them, the preset cumulative path bending energy calculation formula is specifically:

[0062]

[0063] Among them, is the cumulative path bending energy, is the index of the k-th node pair in the first alignment path, is the number of nodes in the first alignment path, is the index difference between the (k + 1)-th node and the k-th node in the intercepted lightning strike distorted feature vector sequence, is the index difference between the (k + 1)-th node and the k-th node in the standard trajectory sequence of the lightning strike feature.

[0064] Among them, the chronological order constraint condition, the specific formula is as follows:

[0065]

[0066] It should be noted that the chronological order constraint condition ensures that the interpolation result conforms to the physical causality by forcing the voltage dip time to be later than the lightning strike marking time , avoiding waveform distortion caused by time sequence misalignment. This constraint condition can effectively improve the interpolation accuracy, ensuring that the restored voltage curve is smooth and conforms to the standard trajectory. For example, strictly aligning the voltage waveform during the lightning strike period to avoid unreasonable interpolation results caused by time axis misalignment. At the same time, the chronological order constraint also enhances the robustness of the system, which can automatically eliminate abnormal paths (such as time sequence misalignment), reducing incorrect interpolation caused by data noise or algorithm errors. For example, if the lightning strike marking time is = 100ms, the voltage dip time after interpolation = 105ms, which meets the chronological constraint; while if = 95ms, then this path will be automatically excluded to avoid introducing unreasonable data filling results. In addition, the chronological constraint can also effectively prevent timing logic conflicts caused by lightning interference, ensuring that the data filling results conform to the actual physical laws. For example, in a lightning interference scenario, the chronological order of the voltage drop section must be consistent with the lightning event; otherwise, it may lead to misjudgment of the device state or logic conflicts. Through the chronological constraint, the system can maintain high precision and high reliability in a complex electromagnetic interference environment, significantly reducing the risk of misoperation and enhancing the stability and security of the power grid operation.

[0067] Among them, the topological logic constraint conditions are specifically formulated as follows:

[0068]

[0069] Among them, is the time point of the voltage drop of the second aligned path, is the time point of the lightning strike mark, is the feature vector of device i, is the feature vector of device j, is the topological matrix element, is the preset topological constraint threshold.

[0070] It should be noted that the topological logic constraint conditions ensure that the device state conforms to the topological interlock logic by forcing the feature differences of directly connected devices to be minimized, avoiding logic conflicts caused by data filling. This constraint condition can effectively guarantee the consistency of the device state, ensuring that the data filling results are consistent with the actual operation state of the power grid. For example, if the bus voltage is 3.0 kV, then the breaker state is forced to be closed to avoid mis-tripping operations. At the same time, the topological logic constraint can also enhance the system security by verifying whether the data filling results conform to the device safe operation rules, reducing the risk of device damage or accidents. In addition, the topological logic constraint can also enhance the credibility of the data filling results by verifying the consistency between the device state and the topological matrix, avoiding logic errors caused by data filling. For example, after data filling, the device state is consistent with the topological matrix, ensuring the correctness of the system operation logic. Through the topological logic constraint, the system can maintain high precision and high reliability in a complex power grid topology environment, significantly reducing the risk of misoperation and enhancing the stability and security of the power grid operation. This constraint condition can also adapt to application scenarios with different voltage levels and device types, having strong scalability and adaptability, providing strong technical support for the anti-misoperation control of smart grids.

[0071] S4. Build an anti-misoperation determination model based on the feature vector after data filling and output anti-misoperation instructions in real time.

[0072] In this example, an anti-misjudgment model is constructed based on the feature vectors after filling in missing values, and anti-misoperation instructions are output in real time. Specifically:

[0073] Obtain historical normal operation device data and perform feature extraction to obtain the first feature vector;

[0074] Perform clustering analysis on the first feature vector to obtain the first pattern template library;

[0075] Train the feature vectors after filling in missing values and the first pattern template library based on the self-supervised contrast learning algorithm to obtain the anti-misjudgment model;

[0076] Identify the real-time data of the power grid device to be judged based on the anti-misjudgment model to obtain anti-misoperation instructions.

[0077] It should be noted that historical normal operation device data is obtained and feature extraction is performed to obtain the first feature vector. By analyzing the historical data of normal devices under different working conditions, their key features are extracted. These features can provide the typical patterns of devices during normal operation, providing a benchmark for subsequent misjudgment identification and helping to distinguish normal operations from potential abnormal states.

[0078] Then, based on these first feature vectors, clustering analysis is performed to obtain the first pattern template library. Through clustering analysis, the normal operation modes of devices can be divided into different categories, and each category represents the operation mode of the device under specific conditions. This template library provides rich reference data for the anti-misjudgment model, enabling it to identify the normal behaviors of devices in different operating states and enhancing the model's identification ability.

[0079] On this basis, the feature vectors after filling in missing values and the first pattern template library are combined and trained through the self-supervised contrast learning algorithm to obtain the anti-misjudgment model. Self-supervised contrast learning is a data learning method without artificial labels. It automatically discovers effective feature representations by comparing the similarities and differences between different feature vectors and uses them for anti-misjudgment. The advantage of this method is that it can make full use of the historical data of devices without a large amount of labeled data, learn high-quality feature representations, and thus improve the generalization ability and accuracy of the model.

[0080] Finally, the real-time data of the power grid device to be judged is input into the anti-misjudgment model for identification to obtain anti-misoperation instructions. Based on the output of the model, the system can judge the operating state of the power grid device in real time and generate corresponding anti-misoperation instructions according to the judgment results, such as alarming, adjusting device parameters, or starting standby devices, etc., so as to effectively avoid misjudgment caused by lightning interference and ensure the stability and safety of the power grid.

[0081] In addition, the core reason why the feature vectors after data repair can be effectively applied to the anti-misoperation of the power grid is that they can restore the true state of the equipment by dynamically repairing the distorted data caused by disturbances such as lightning strikes, and achieve accurate risk decision-making through a multi-modal judgment mechanism. Specifically, when power grid equipment is under transient disturbances such as lightning strikes, the original data collected will be distorted due to sudden increases in noise, waveform distortion, or data loss. Feature vectors directly generated based on such distorted data may misrepresent the equipment state, which may lead to the anti-misoperation system mistakenly allowing high-risk operations. Data repair techniques (such as DTW path screening) repair the distorted parts in the feature vectors by locating the distorted areas, matching the templates of historical normal data, and replacing the outliers, making them closer to the true physical laws. For example, the sudden drop in current caused by lightning strikes can be replaced with the normal current value under similar working conditions, thus eliminating the interference and misguidance. After the repaired feature vectors are input into the anti-misoperation judgment model, by calculating the similarity with the normal template library or analyzing the reconstruction error of unsupervised models (such as autoencoders), it is determined whether the current state deviates from the safe range; supervised models (such as LSTM, random forest) further classify the risk levels and generate operation instructions in combination with the expert rule library (such as "switching off is prohibited under live conditions"). If it is determined to be abnormal, the system immediately locks the dangerous operation (such as locking the switching off of the live switch) and triggers an alarm or redundant protection, otherwise it allows the execution. The technical advantages of this process are reflected in three aspects: First, it eliminates the interference of data distortion on state judgment and avoids misjudgments caused by "false normal" or "false abnormal"; second, it improves the adaptability of the model to different operating scenarios by covering all working conditions of the equipment (such as light load, heavy load, aging) through a multi-modal template library; third, the dynamic closed-loop mechanism (real-time repair - judgment - feedback) ensures that the system can still maintain high reliability in complex environments such as frequent lightning strikes and load fluctuations. For example, when a substation plans to disconnect the bus switch, the feature vectors after data repair accurately reflect that the equipment is still in a live state. Based on this, the anti-misoperation model matches the "switching off is prohibited under live conditions" rule, locks the switching off instruction and gives an alarm, while the distorted data without data repair may be misjudged as "no current" and allow the operation, resulting in a major accident. Therefore, the feature vectors after data repair essentially deeply couple data quality with judgment logic, and build a safe closed-loop from data to operation through "removing false and retaining true" data repair and "multi-dimensional verification" intelligent decision-making, ultimately achieving the leap of the power grid anti-misoperation system from "passive response" to "active immunity".

[0082] Embodiment 2, Figure 2 The dynamic anti-misoperation system for power grid equipment based on intelligent topology according to the present invention is given, including a data acquisition module, a distortion identification module, a vector repair module, and an anti-misoperation judgment module:

[0083] The data acquisition module is used to construct a topological matrix of the power grid equipment and fuse it with the equipment parameters to obtain anti-misoperation feature vectors;

[0084] A distortion recognition module for performing time-domain analysis on the anti-error feature vector to identify the lightning strike distortion feature vector;

[0085] A vector complementing module for complementing the lightning strike distortion feature vector based on DTW path screening;

[0086] An anti-error determination module for constructing an anti-error determination model based on the complemented feature vector and real-time outputting an anti-error operation instruction.

[0087] Embodiment 3 Figure 3 This is the DTW signal alignment path broken line graph regarding DTW path screening in the embodiment provided by the present invention. The standard trajectory sequence (solid line) presents the ideal lightning strike waveform characteristics: it steeply rises to a peak value of 1.2 pu at the nanosecond level in the initial 0 - 20 ms, decays according to the exponential law e−8t in the middle period of 20 - 60 ms, and maintains a residual oscillation below 0.2 pu in the later period of 60 - 100 ms, fully reproducing the IEC standard double-exponential model, and its waveform data is derived from laboratory high-voltage test calibration. The lightning strike distortion feature vector sequence (dashed line) reflects typical industrial site interferences: there is a 5 ms time shift in the overall waveform (corresponding to a 50-meter cable transmission delay), the amplitude decays by 40% to 0.72 pu in the 30 - 50 ms section due to CT magnetic saturation, and Gaussian white noise with a standard deviation of 0.3 is superimposed in the 70 - 90 ms section (signal-to-noise ratio 14.5 dB), comprehensively characterizing the sensor nonlinear distortion and electromagnetic interference effects.

[0088] All the above formulas are dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation, and the preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0089] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0090] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0091] In addition, the functional modules in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0092] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

[0093] Finally: The above description is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for dynamically preventing misoperation of power grid equipment based on intelligent topology, characterized in that It includes the following steps: Construct a topological matrix of grid equipment and fuse it with equipment parameters to obtain an anti-error feature vector; Conduct time-domain analysis on the anti-error feature vector to identify lightning strike distortion feature vectors; Perform time window truncation on the lightning strike distortion feature vectors to obtain a sequence of truncated lightning strike distortion feature vectors; Construct a standard trajectory sequence of lightning strike features, and use the DTW algorithm to perform path offset matching between the sequence of truncated lightning strike distortion feature vectors and the standard trajectory sequence of lightning strike features to obtain a number of first alignment paths; Screen the number of first alignment paths to obtain an optimal alignment path; Complement the lightning strike distortion feature vectors based on the optimal alignment path; Construct an anti-error determination model based on the complemented feature vectors and output anti-error operation instructions in real time.

2. The dynamic anti-misoperation method for grid equipment startup based on intelligent topology according to claim 1, wherein The construction of the topological matrix of grid equipment and the fusion with equipment parameters to obtain an anti-error feature vector is specifically as follows: Obtain real-time equipment parameter data of the target area and obtain a topological matrix based on frequency domain coupling degree analysis; Perform feature fusion on the real-time equipment parameter data and the topological matrix to obtain an anti-error feature vector.

3. The method for starting dynamic error prevention of power grid equipment based on intelligent topology according to claim 2, characterized in that, The time-domain analysis of the anti-error feature vector to identify lightning strike distortion feature vectors is specifically as follows: Perform wavelet packet decomposition on the anti-error feature vector to extract the high-frequency band energy ratio feature; Obtain the energy eigenvalue of the frequency band reference, and compare and analyze the ratio of the high-frequency band energy ratio feature to the energy eigenvalue of the frequency band reference with a first threshold to obtain a lightning strike distortion feature vector.

4. The method for starting dynamic error prevention of grid equipment based on intelligent topology according to claim 3, characterized in that The screening of the number of first alignment paths to obtain an optimal alignment path is specifically as follows: Calculate the local slope of each node and its predecessor node in the number of first alignment paths; Calculate the cumulative path bending energy of each first alignment path based on a preset cumulative path bending energy calculation formula; Eliminate the first alignment paths whose cumulative path bending energy does not meet the preset constraint conditions to obtain a second alignment path; Perform secondary screening on the second alignment path based on time sequence constraint conditions and topological logic constraint conditions to obtain an optimal alignment path.

5. The method for starting dynamic error prevention of power grid equipment based on intelligent topology according to claim 4, wherein, The complementing of the lightning strike distortion feature vectors based on the optimal alignment path is specifically as follows: Analyze the optimal alignment path to obtain the time index of the distorted signal and the time index of the standard signal; Map the time points of the distorted signal to the time points of the standard signal and synchronize them; Correct the amplitude of the distorted signal and replace it with the amplitude of the corresponding standard signal to obtain the complemented feature vector.

6. The method for starting dynamic error prevention of power grid equipment based on intelligent topology according to claim 5, characterized in that, The obtaining of the topological matrix based on frequency domain coupling degree analysis is specifically as follows: wherein, is the frequency-domain coupling degree between device i and device j, is the frequency, is the lightning strike sensitive frequency band, is the cross-spectral density between device i and device j, is the lightning strike interference spectrum eigenvalue, is a constant, is the cross-spectral density of device i, is the cross-spectral density of device j.

7. The method for starting dynamic error prevention of power grid equipment based on intelligent topology according to claim 6, wherein The preset cumulative path bending energy calculation formula is specifically as follows: Among them, is the cumulative path bending energy, is the index of the k-th node pair in the first alignment path, is the number of nodes in the first alignment path, is the index difference between the (k + 1)-th node and the k-th node on the intercepted lightning strike distortion feature vector sequence, is the index difference between the (k + 1)-th node and the k-th node on the standard trajectory sequence of the lightning strike feature.

8. The method for starting dynamic error prevention of grid equipment based on intelligent topology according to claim 7, wherein, The secondary screening of the second alignment path based on time sequence constraint conditions and topological logic constraint conditions to obtain an optimal alignment path is specifically as follows: The time sequence constraint conditions are specifically as follows: The topological logic constraint conditions are specifically as follows: Among them, is the time point of the second aligned path voltage dip, is the time point of the lightning strike mark, is the feature vector of device i, is the feature vector of device j, is the topological matrix element, is the preset topological constraint threshold.

9. The power grid equipment startup dynamic error prevention system based on intelligent topology is applied to the power grid equipment startup dynamic error prevention method based on intelligent topology according to any one of claims 1-8, and is characterized in that It includes a data acquisition module, a distortion identification module, a vector complementing module, and an anti-error determination module: The data acquisition module is used to construct a topological matrix of grid equipment and fuse it with equipment parameters to obtain an anti-error feature vector; The distortion identification module is used to conduct time-domain analysis on the anti-error feature vector to identify lightning strike distortion feature vectors; A vector complementation module, which is used to complement the lightning strike distortion feature vector based on the DTW path screening; An anti-misjudgment module, which is used to construct an anti-misjudgment model based on the complemented feature vector and output an anti-misoperation instruction in real time.

Images