Power transmission and transformation project intelligent index cross checking method based on intelligent management and control platform

Abstract

The invention discloses a power transmission and transformation project intelligent index cross checking method based on an intelligent management and control platform, and relates to the technical field of data correction, and the method comprises the following steps: obtaining grounding characteristic data of a lightning stroke region, and screening the grounding characteristic data; acquiring real-time lightning stroke data, and reconstructing an overvoltage waveform in combination with the screened grounding characteristic data; the insulator flashover probability is predicted according to the reconstructed overvoltage waveform, and a dynamic flashover probability curve is output; performing feature extraction on the dynamic flashover probability curve, and performing flashover risk assessment according to the extracted features; based on a regression analysis method, establishing a mapping relation model between the flashover risk and the load fluctuation correction value, and correcting the load fluctuation data according to the mapping relation model; according to the method and the device, the load fluctuation data is dynamically corrected through cross checking, so that the accuracy of receiving the data is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data correction, and more specifically, to a method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform. Background Art

[0002] With the rapid development of information technology and the in-depth advancement of smart grid construction, the intelligence level of power transmission and transformation projects has become a key indicator of the modernization of power systems. As a core component of smart grids, smart management and control platforms integrate advanced artificial intelligence, big data, and the Internet of Things (IoT) technologies to enable comprehensive monitoring, intelligent analysis, and efficient management of the operating status of power transmission and transformation equipment. Based on this, a cross-verification method for intelligent indicators of power transmission and transformation projects, based on the smart management and control platform, has emerged, providing strong support for improving the operational efficiency and safety of power transmission and transformation projects.

[0003] By integrating security systems such as video surveillance, facial recognition, and access control, combined with big data analytics, the intelligent management and control platform enables efficient management of equipment and assets within the power transmission and transformation project area. The platform monitors equipment operating status in real time, promptly identifying and warning of potential faults. It also supports the integration of functions such as equipment maintenance, alarm processes, and asset architecture, significantly enhancing the intelligence of operations and maintenance management. Furthermore, the platform includes energy management and environmental monitoring systems to ensure efficient resource utilization and environmental comfort, further enhancing the sustainability of the power transmission and transformation project.

[0004] Based on the smart management and control platform, a cross-verification method for intelligent indicators of power transmission and transformation projects has been implemented. This method utilizes the extensive real-time data provided by the platform and combines it with cross-analysis principles to conduct in-depth research and cross-validation of various intelligent indicators of power transmission and transformation projects. As a method for studying the relationship between categorical and noncategorical data, cross-analysis can reveal the inherent connections and underlying patterns between different indicators, thus providing a scientific basis for accurate assessment of the level of intelligence.

[0005] For example, the invention patent with announcement number CN112927063A discloses a financial information cross-verification method. First, a cross-matching model and the data items required for cross-validation are configured for the model. The data items are then grouped, with the majority of the verification data used as an evaluation set and a smaller portion used as a validation set. The validation set data undergoes manual inspection, with the data information results roughly estimated using relevant expressions and compared with the verification results of the evaluation set data. The data inspection and comparison are run based on relevant industry inspection indicators and rules, and output two results: "consistent" and "inconsistent." The "inconsistent" results output by the data inspection process are automatically recorded. This financial information cross-verification method significantly reduces the workload of manual verification by splitting large amounts of financial information into evaluation and validation sets for computer and manual cross-verification, facilitating more comprehensive and accurate verification with less manual effort.

[0006] For example, the invention patent with announcement number: CN112053238A discloses a method, device and system for cross-verification of information, which obtains the data information filled in by the verification user, collects valid information in the data information according to a preset verification mode, cross-verifies the valid information, and obtains the verification result corresponding to the cross-verification. If the verification result is passed, it means that the verification is successful. If the verification result is failed, the operation information feedback by the verification user is obtained. If the feedback operation information of the verification user is to re-fill in the data information, the data information is re-obtained and cross-verified. If the feedback operation information of the verification user is to fill in a supplementary answer or submit information, a risk prompt is generated for the user in the background, thereby realizing automatic cross-verification of data information, effectively avoiding the problems of slow manual verification and low accuracy, and effectively improving the efficiency of cross-verification.

[0007] The above disclosed technical solutions have at least the following technical problems: Traditional intelligent indicator cross-checking methods for power transmission and transformation projects based on smart management and control platforms fail to effectively address the attenuation and non-stationary characteristics of high-frequency signals when reconstructing overvoltage waveforms due to the effects of lightning strikes, resulting in waveform distortion and inaccurate data received by the system. This present invention proposes a solution to this problem. Summary of the Invention

[0008] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent indicator cross-calibration method for power transmission and transformation projects based on a smart management and control platform. Through cross-calibration, load fluctuation data is dynamically corrected to improve the accuracy of received data.

[0009] To achieve the above object, the present invention provides the following technical solutions: An intelligent indicator cross-check method for power transmission and transformation projects based on a smart management and control platform includes the following steps: obtaining grounding characteristic data in the lightning strike area and screening the grounding characteristic data; obtaining real-time lightning strike data and reconstructing the overvoltage waveform in combination with the screened grounding characteristic data; predicting the insulator flashover probability based on the reconstructed overvoltage waveform and outputting a dynamic flashover probability curve; extracting features from the dynamic flashover probability curve and assessing the flashover risk based on the extracted features; establishing a mapping relationship model between the flashover risk and the load fluctuation correction value based on the regression analysis method, and correcting the load fluctuation data based on the mapping relationship model.

[0010] In a preferred embodiment, the grounding feature data of the lightning strike area is obtained and the grounding feature data is screened, specifically as follows: historical grounding system maintenance data of the area to be tested is obtained, and the grounding feature data at the time of lightning strike is extracted; the grounding feature data is preliminarily screened through variance analysis, and the grounding feature data with variance approaching zero is eliminated, and the top N grounding feature data are retained; after the grounding feature data is preliminarily screened, a predefined key feature list is checked, and key dynamic features are retained through physical constraint screening; the grounding features retained after screening are merged with the key dynamic features retained through physical constraint screening to form a grounding feature set.

[0011] In a preferred embodiment, the real-time lightning strike data is obtained and the overvoltage waveform is reconstructed in combination with the screened grounding feature data, specifically as follows: real-time lightning strike data is obtained, and the real-time lightning strike data includes lightning strike location and lightning strike current data; according to the lightning strike location and the grounding feature set, the grounding feature data of the corresponding area is matched; according to the matched grounding feature data and lightning strike current data, a data-driven model of the overvoltage waveform is established to generate a reference overvoltage waveform; based on the adversarial generative network, the lightning strike data, the grounding feature data and the reference overvoltage waveform are input to generate an initial overvoltage waveform; the high-frequency component of the initial overvoltage waveform is reconstructed, and the initial overvoltage waveform is dynamically corrected.

[0012] In a preferred embodiment, the high-frequency component of the initial overvoltage waveform is reconstructed and the initial overvoltage waveform is dynamically corrected as follows: the initial overvoltage waveform is obtained, and the initial overvoltage waveform is fast Fourier transformed, the spectrum is output, and the high-frequency component is obtained; the high-frequency component is decomposed at multiple scales, and the overvoltage waveform is subjected to time-frequency analysis to extract the high-frequency signal; the high-frequency signal is filtered, and the high-frequency signal is decomposed based on the EMD method to obtain different IMFs, and the high-frequency energy proportions of different IMFs are output; the IMF components that mainly contribute are screened out according to the high-frequency energy proportions of different IMFs; the screened IMF components are superimposed back on the low-frequency components to obtain the corrected overvoltage waveform.

[0013] In a preferred embodiment, the insulator flashover probability is predicted based on the reconstructed overvoltage waveform and a dynamic flashover probability curve is output, specifically as follows: the peak voltage of the reconstructed overvoltage waveform is extracted, and the initial equivalent action time is output based on the waveform integration method; the flashover probability curve is output based on the exponential decay model according to the peak voltage; the flashover probability curve is reversed based on the Weibull distribution model through the flashover time sensitivity to obtain the reversed equivalent action time; the flashover probability curve is dynamically corrected according to the reversed equivalent action time and the initial equivalent action time, and the dynamic flashover probability curve is output.

[0014] In a preferred embodiment, the flashover probability curve is dynamically corrected according to the equivalent action time obtained by inverse calculation and the initial equivalent action time, and a dynamic flashover probability curve is output, specifically as follows: according to the inverse equivalent action time and the initial equivalent action time, a change ratio of the inverse equivalent action time relative to the initial equivalent action time is obtained; the inverse equivalent action time is corrected according to the change ratio to obtain a modified Weibull distribution model; the flashover probability curve output by the modified Weibull distribution model is corrected by spline interpolation and exponential weighted smoothing method, and the dynamic flashover probability curve is output.

[0015] In a preferred embodiment, the feature extraction of the dynamic flashover probability curve and the flashover risk assessment are performed based on the extracted features, specifically as follows: time series data are extracted from the dynamic flashover probability curve; time domain analysis is performed on the time series data to extract time domain features; and the time domain features are input based on a logistic regression equation to obtain a flashover risk assessment result.

[0016] In a preferred embodiment, the load fluctuation data is corrected according to the mapping relationship model as follows: reverse deduction is performed according to the mapping relationship model to solve the load fluctuation data when the flashover risk probability threshold and the current flashover risk probability are satisfied, that is, the corrected load fluctuation.

[0017] The technical effects and advantages of the intelligent indicator cross-check method for power transmission and transformation projects based on the smart management and control platform of the present invention are as follows: 1. This invention integrates grounding feature data, real-time lightning strike data, and environmental parameters in the lightning strike area to construct a closed-loop analysis system for dynamic overvoltage waveform reconstruction and flashover probability prediction. Variance analysis and physical constraint screening techniques are used to effectively eliminate redundant data (such as invalid features with variance approaching zero) while retaining key dynamic features (such as impulse impedance and high-frequency component energy ratio), significantly improving the representativeness and reliability of the feature set. The KD tree nearest neighbor search algorithm is used to match the lightning strike location with the grounding feature, and the generative adversarial network (GAN) is used to reconstruct the overvoltage waveform of the high-frequency component. This solves the waveform distortion problem caused by high-frequency signal attenuation in traditional methods, reduces the overvoltage waveform restoration error to less than 5%, and provides high-fidelity input for flashover probability prediction.

[0018] 2. This invention's flashover probability prediction framework, based on the Weibull distribution model, addresses the inability of traditional static models to adapt to transient operating conditions by inferring equivalent action time and implementing a dynamic correction mechanism. By incorporating spline interpolation and exponentially weighted smoothing into the probability curve for dynamic correction, the mean square error (MSE) of the flashover probability prediction is reduced. The prediction sensitivity is significantly improved, particularly during the initial stages of a lightning strike, enabling early warning signals to be triggered, creating a critical response window for grid dispatch.

[0019] 3. This invention achieves risk-driven dynamic load optimization by establishing a nonlinear mapping relationship model between flashover risk and load fluctuation. Using a reverse derivation algorithm to solve for load correction values that meet safety thresholds, the flashover risk probability threshold can be compressed without interrupting power supply, while ensuring that load fluctuations are controlled within a ±2% range. This method breaks through the traditional passive management model of "fixed threshold + manual intervention" and improves the operational resilience of the power grid in extreme weather conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The figure is a flow chart of a method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform according to the present invention. DETAILED DESCRIPTION

[0021] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0022] Example 1, Figure 1 The present invention provides a method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform, which includes the following steps: S1, obtaining grounding characteristic data of the lightning strike area and screening the grounding characteristic data.

[0023] In this embodiment, grounding characteristic data of the lightning strike area is obtained and the grounding characteristic data is screened as follows: Obtain historical grounding system maintenance data for the area to be tested, and extract grounding characteristic data at the time of lightning strike based on the historical grounding system maintenance data, wherein the grounding characteristic data includes but is not limited to grounding resistance, soil resistivity, and impulse impedance; The grounding characteristic data are preliminarily screened by variance analysis, and the grounding characteristic data with variance close to zero are eliminated, and the top N grounding characteristic data are retained, including the grounding bodies with fixed burial depths; After preliminary screening of ground feature data, a predefined list of key features is checked to retain key dynamic features through physical constraint screening; The grounding features retained after screening are merged with the key dynamic features retained by physical constraint screening to form a grounding feature set.

[0024] It should be noted that the final screened feature set is verified to ensure that all key dynamic features are included, while redundant or uninformative static features are eliminated, and the final feature set is used for subsequent model construction or further data analysis tasks.

[0025] S2, obtains real-time lightning strike data and reconstructs the overvoltage waveform in combination with the filtered grounding characteristic data.

[0026] In this embodiment, real-time lightning strike data is acquired and combined with filtered grounding characteristic data to reconstruct the overvoltage waveform, as follows: Acquire real-time lightning strike data, including lightning strike location, lightning strike current data, wavefront time, and return stroke number; Preprocess lightning strike data, using wavelet transform or low-pass filter to eliminate high-frequency interference, remove unreasonable data based on the physical constraints of lightning current, and synchronize the timestamps of multi-source data; According to the nearest neighbor search algorithm based on the KD tree at the lightning strike location, the grounding feature set is matched to obtain the grounding feature data of the corresponding area; Based on the matched grounding characteristic data and lightning current data, a data-driven model of the overvoltage waveform is established to generate a reference overvoltage waveform; Based on the adversarial generative network, lightning strike data, grounding characteristic data and reference overvoltage waveform are input to generate the initial overvoltage waveform; Wavelet transform and Fourier transform are used to analyze and reconstruct the high-frequency components of the initial overvoltage waveform, and the initial overvoltage waveform is dynamically corrected.

[0027] In this embodiment, wavelet transform and Fourier transform are used to analyze and reconstruct the high-frequency component of the initial overvoltage waveform, and the initial overvoltage waveform is dynamically corrected, as follows: Obtain the initial overvoltage waveform, perform fast Fourier transform on the initial overvoltage waveform, output the spectrum, and identify the high-frequency components; The high-frequency components are decomposed into multiple scales through discrete wavelet transform, the optimal wavelet basis function is selected, the overvoltage waveform is subjected to time-frequency analysis, and the high-frequency signal is extracted to ensure that the time-varying characteristics of the high-frequency components are captured; Filter high-frequency signals, filter low-frequency interference, decompose high-frequency signals based on the empirical mode decomposition (EMD) method, obtain different intrinsic mode components (IMFs), and output different high-frequency energy proportions of IMFs; Filter out the main contributing IMF components based on the high-frequency energy proportion of different IMFs; The filtered IMF components are superimposed back onto the low-frequency components to obtain the corrected overvoltage waveform.

[0028] The calculation formula of the eigenmode component is as follows:

[0029] The calculation formula for the high-frequency energy ratio of IMF is as follows:

[0030] Where: is the high frequency energy proportion of the eigenmode component, is the eigenmode component of the high-frequency signal, are high-frequency components of different scales. Each IMF represents the oscillation mode of the signal in different frequency bands. is the number of eigenmode components, It is the remaining trend component, that is, the part of the signal that can no longer be decomposed, usually a low-frequency or DC trend component.

[0031] The calculation formula for superimposing the filtered high-frequency IMF components back to the low-frequency components is as follows:

[0032] Where: is the corrected overvoltage waveform, is the low-frequency component of the initial overvoltage signal, is the filtered high-frequency component.

[0033] S3, predict the insulator flashover probability based on the reconstructed overvoltage waveform and output a dynamic flashover probability curve.

[0034] In this embodiment, the insulator flashover probability is predicted based on the reconstructed overvoltage waveform, and a dynamic flashover probability curve is output, as follows: Extract the peak voltage of the reconstructed overvoltage waveform and output the initial equivalent action time based on the waveform integration method; Output flashover probability curve based on exponential decay model according to peak voltage; According to the flashover probability curve, the equivalent action time is obtained by reverse deduction based on the flashover time sensitivity and Weibull distribution model; The flashover probability curve is dynamically corrected according to the equivalent action time obtained by reverse calculation and the initial equivalent action time, and a dynamic flashover probability curve is output.

[0035] The calculation formula for the equivalent action time output by the waveform integration method is as follows:

[0036] The calculation formula of the exponential decay model is as follows:

[0037] The calculation formula of the Weibull distribution model is as follows:

[0038] Where: is the flashover probability, is the peak voltage of the reconstructed overvoltage waveform, is the voltage level at which 50% of flashovers occur, is a constant control factor, is the equivalent action time, is the characteristic time constant, reflecting the flashover time sensitivity, is the shape parameter, which controls the steepness of the curve. is the corrected overvoltage waveform.

[0039] In this embodiment, the flashover probability curve is dynamically corrected according to the equivalent action time obtained by reverse calculation and the initial equivalent action time, and a dynamic flashover probability curve is output, which is specifically as follows: According to the equivalent action time obtained by reverse calculation and the initial equivalent action time, a change ratio of the equivalent action time obtained by reverse calculation to the initial equivalent action time is obtained; According to the equivalent action time obtained by back-calculation in the modified Weibull distribution model of the change ratio, the modified Weibull distribution model is obtained; The flashover probability curve outputted by the modified Weibull distribution model has a large deviation from the original curve. It is corrected by spline interpolation and exponential weighted smoothing method to avoid curve mutation and ensure natural transition, and output a dynamic flashover probability curve.

[0040] The calculation formula for the change ratio is as follows:

[0041] The calculation formula of the equivalent action time obtained by correcting the back-calculation is as follows:

[0042] Where: is the rate of change, is the initial equivalent action time, is the equivalent action time obtained by reverse calculation, is the corrected equivalent action time of the backpropagation.

[0043] It should be noted that if the change ratio is greater than 1, it means that the actual action time is longer than the initial estimate, which means that the flashover probability needs to be increased; if the change ratio is less than 1, it means that the actual action time is shorter than the initial estimate, which means that the flashover probability needs to be reduced.

[0044] S4, extracting features from the dynamic flashover probability curve and performing flashover risk assessment based on the extracted features.

[0045] In this embodiment, feature extraction is performed on the dynamic flashover probability curve, and flashover risk assessment is performed based on the extracted features, as follows: Extracting time series data from the dynamic flashover probability curve, performing time domain analysis on the time series data, and extracting time domain features, wherein the time domain features include flashover probability peak value, rise rate, average flashover probability, and probability volatility; Perform frequency domain analysis on time domain features to detect periodic characteristics, output the main frequency component and amplitude, and determine whether there is harmonic interference or environmental periodic influence. If the amplitude of a high-frequency component is large, further analysis of external disturbance factors (such as lightning and grid harmonics) may be required. The flashover risk assessment results are obtained by inputting the time domain characteristics into the logistic regression equation.

[0046] The calculation formula for flashover risk assessment is as follows:

[0047] Where: is the flashover risk assessment result, 、 、 、 and is the regression coefficient, is the peak flashover probability, is the rate of ascent, is the average flashover probability, is the flashover probability volatility, is the flashover risk classification label, .

[0048] The advantages of comprehensively evaluating flashover risk by using flashover probability peak, rise rate, average flashover probability, and probability volatility are further explained: Traditional flashover risk assessments often rely on a single statistic, such as average flashover probability or maximum overvoltage. This approach may overlook the impact of instantaneous fluctuations on flashover risk. By introducing flashover probability peak, rise rate, average flashover probability, and probability fluctuation, flashover risk can be assessed from multiple perspectives. Flashover probability peak: This indicates the maximum flashover probability in the overvoltage waveform at a specific moment, reflecting the maximum risk to the power grid in an emergency. This can help identify possible "high-risk" periods in the system. Rise rate: describes the rate at which the overvoltage waveform rises, reflecting the speed at which the voltage changes in the power system. A fast rise rate may increase the instantaneous risk of flashover, especially under rapidly fluctuating load conditions, where the risk of flashover increases rapidly. Average flashover probability: By calculating the average flashover probability of the overvoltage waveform within a certain time window, the risk level of the power grid in long-term operation can be assessed. This helps in long-term stability analysis and identifies potential risk factors. Probabilistic volatility: This reflects the fluctuation of flashover risk over time. Larger volatility indicates unstable grid overvoltage and a greater risk of sudden overvoltage events, while smaller volatility indicates relatively stable grid operation and a lower risk of flashover events.

[0049] S5. Based on the regression analysis method, a mapping relationship model between the flashover risk and the load fluctuation correction value is established, and the load fluctuation data is corrected according to the mapping relationship model.

[0050] In this embodiment, based on the regression analysis method, a mapping relationship model between the flashover risk and the load fluctuation correction value is established, and the load fluctuation data is corrected according to the mapping relationship model, as follows: According to the mapping relationship model, reverse deduction is performed to solve the load fluctuation data when the flashover risk probability threshold and the current flashover risk probability are met, and the load fluctuation is corrected.

[0051] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0052] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0053] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0054] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0055] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0056] Finally: 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 in the scope of protection of the present invention.

Claims

1. The intelligent indicator cross-check method for power transmission and transformation projects based on the smart management and control platform is characterized by: The steps include: Obtaining grounding characteristic data in the lightning strike area and screening the grounding characteristic data; Acquire real-time lightning strike data and reconstruct the overvoltage waveform by combining it with filtered grounding characteristic data; Predict the insulator flashover probability based on the reconstructed overvoltage waveform and output a dynamic flashover probability curve; Extract features from the dynamic flashover probability curve and perform flashover risk assessment based on the extracted features; Based on the regression analysis method, a mapping relationship model between flashover risk and load fluctuation correction value is established, and the load fluctuation data is corrected according to the mapping relationship model.

2. The intelligent indicator cross-check method for power transmission and transformation engineering based on the smart management and control platform according to claim 1 is characterized in that: The method of obtaining the grounding characteristic data of the lightning strike area and screening the grounding characteristic data is as follows: Obtain historical grounding system maintenance data for the area to be tested and extract grounding characteristic data during lightning strikes; The grounding characteristic data are preliminarily screened by variance analysis, the grounding characteristic data with variance close to zero are eliminated, and the top N grounding characteristic data are retained; After preliminary screening of ground feature data, a predefined list of key features is checked to retain key dynamic features through physical constraint screening; The grounding features retained after screening are merged with the key dynamic features retained by physical constraint screening to form a grounding feature set.

3. The intelligent indicator cross-check method for power transmission and transformation engineering based on the smart management and control platform according to claim 2 is characterized in that: The real-time lightning strike data is obtained and the overvoltage waveform is reconstructed in combination with the filtered grounding characteristic data as follows: Acquiring real-time lightning strike data, wherein the real-time lightning strike data includes lightning strike location and lightning strike current data; Matching the grounding feature data of the corresponding area according to the lightning strike location and the grounding feature set; Based on the matched grounding characteristic data and lightning current data, a data-driven model of the overvoltage waveform is established to generate a reference overvoltage waveform; Based on the adversarial generative network, lightning strike data, grounding characteristic data and reference overvoltage waveform are input to generate the initial overvoltage waveform; The high-frequency component of the initial overvoltage waveform is reconstructed and the initial overvoltage waveform is dynamically corrected.

4. The method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform according to claim 3 is characterized in that: The high-frequency component of the initial overvoltage waveform is reconstructed and the initial overvoltage waveform is dynamically corrected as follows: Obtaining an initial overvoltage waveform, performing a fast Fourier transform on the initial overvoltage waveform, outputting a spectrum, and obtaining a high-frequency component; Decompose the high-frequency components into multiple scales and perform time-frequency analysis on the overvoltage waveform to extract the high-frequency signal; Filter the high-frequency signal, decompose it based on the EMD method, obtain different IMFs, and output the high-frequency energy ratio of different IMFs; Filter out the main contributing IMF components based on the high-frequency energy proportion of different IMFs; The filtered IMF components are superimposed back onto the low-frequency components to obtain the corrected overvoltage waveform.

5. The intelligent indicator cross-check method for power transmission and transformation engineering based on the smart management and control platform according to claim 4 is characterized in that: The insulator flashover probability is predicted based on the reconstructed overvoltage waveform, and a dynamic flashover probability curve is output, as follows: Extract the peak voltage of the reconstructed overvoltage waveform and output the initial equivalent action time based on the waveform integration method; Output flashover probability curve based on exponential decay model according to peak voltage; According to the flashover probability curve, the flashover time sensitivity is reversed based on the Weibull distribution model to obtain the reverse equivalent action time; The flashover probability curve is dynamically corrected according to the inverse equivalent action time and the initial equivalent action time, and a dynamic flashover probability curve is output.

6. The method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform according to claim 5 is characterized in that: The flashover probability curve is dynamically corrected according to the inverse equivalent action time and the initial equivalent action time, and a dynamic flashover probability curve is output, which is specifically as follows: According to the reverse equivalent action time and the initial equivalent action time, the change ratio of the reverse equivalent action time to the initial equivalent action time is obtained; According to the change ratio, the equivalent action time is corrected and inverted to obtain the modified Weibull distribution model; The flashover probability curve output by the modified Weibull distribution model is modified by spline interpolation and exponential weighted smoothing method, and a dynamic flashover probability curve is output.

7. The method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform according to claim 6 is characterized in that: The feature extraction of the dynamic flashover probability curve and the flashover risk assessment based on the extracted features are specifically as follows: Extract time series data from dynamic flashover probability curves; Performing time domain analysis on the time series data to extract time domain features; The flashover risk assessment results are obtained based on the time domain characteristics input into the logistic regression equation; The calculation formula for the flashover risk assessment is as follows: Where: is the flashover risk assessment result, 、 、 、 and is the regression coefficient, is the peak flashover probability, is the rate of ascent, is the average flashover probability, is the flashover probability volatility, is the flashover risk classification label, .

8. The method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform according to claim 7 is characterized in that: The load fluctuation data is corrected according to the mapping relationship model as follows: According to the mapping relationship model, reverse deduction is performed to solve the load fluctuation data when the flashover risk probability threshold and the current flashover risk probability are met, that is, the corrected load fluctuation.

9. The method for cross-checking intelligent indicators of power transmission and transformation projects based on a smart management and control platform according to claim 8 is characterized in that: The calculation formula of the change ratio is as follows: The calculation formula of the modified back-estimation equivalent action time is as follows: Where: is the rate of change, is the initial equivalent action time, is the equivalent action time obtained by reverse calculation, is the corrected equivalent action time of the backpropagation.

Images