Intelligent power security situation assessment method based on multi-source sensor network

Abstract

The invention relates to an intelligent power security situation assessment method based on a multi-source sensor network, and the method comprises the steps: obtaining multi-dimensional data from a power system through the multi-source sensor network, and constructing an original data set containing time sequence features; preprocessing and denoising the original data set, and extracting dynamic change characteristics of key parameters to obtain a cleaned parameter change sequence; based on the cleaned parameter change sequence, analyzing an interaction relationship among the key parameters, and obtaining spectrum feature data; judging whether a resonance risk exists in the system or not according to the frequency spectrum characteristic data in combination with a resonance phenomenon theoretical model, and generating a risk assessment report; for the risk assessment report, identifying unstable factors and extracting a high-risk parameter combination, and generating unstable factor distribution feature data; and according to the system stability evaluation index, judging whether the power grid operation state is close to an unstable boundary, and generating a risk early warning signal and a parameter adjustment suggestion.

Description

Technical Field

[0001] The present application relates to the field of electric variable technology, and in particular to an intelligent method for electric power security situation assessment based on a multi-source sensor network. Background Art

[0002] In the field of power system operation and safety management, traditional monitoring and assessment methods are often limited to a single data source and static analysis, making it difficult to fully capture the dynamic changes in system operation. In particular, when faced with complex resonance phenomena, existing methods are insufficient in terms of prediction and early warning capabilities, which can easily lead to potential risks being overlooked. A core challenge in this field is how to accurately identify key factors that may cause instability in the system through multi-dimensional data collection and analysis. The first problem is that the dynamic changes of certain key parameters in the system are difficult to accurately capture, such as the interaction between capacitance and inductance. This interaction will directly affect the stability of the system. If the changing trends of these parameters cannot be grasped in a timely manner, it will be difficult to further determine whether resonance will occur in the system. Once resonance occurs, it may cause abnormal fluctuations in voltage or current, thereby threatening the safe operation of the entire power grid. This logical chain from parameter monitoring to phenomenon prediction constitutes the difficulty of current technological breakthroughs. Summary of the Invention

[0003] In order to solve the problems existing in the above-mentioned prior art, the purpose of this application is to provide an intelligent method for power security situation assessment based on a multi-source sensor network.

[0004] The present application discloses an intelligent method for assessing power security situation based on a multi-source sensor network, comprising: Step S101: By deploying a multi-source sensor network, multi-dimensional data is collected from the power system in real time, covering monitoring points of key parameters such as voltage, current, capacitance, and inductance, and constructing a raw data set containing time series characteristics for subsequent dynamic change analysis; Step S102: Preprocess the raw data set obtained in step S101 using signal processing technology, remove noise interference using filtering methods, extract the dynamic change characteristics of capacitance and inductance values at different time points, and obtain a cleaned parameter change sequence for subsequent time-frequency characteristic analysis; Step S103: Based on the cleaned parameter change sequence obtained in step S102, a time-frequency analysis model based on wavelet transform is used to analyze the interaction relationship between the capacitance and inductance values, calculate the coupling characteristics between the two in the frequency domain, and obtain spectral characteristic data reflecting the interaction strength for subsequent resonance risk assessment; Step S104: Based on the spectrum characteristic data obtained in step S103 and in combination with the theoretical model of the resonance phenomenon, determine whether there are potential conditions for resonance in the system. If the spectrum characteristic data exceeds the resonance threshold preset based on historical data and theoretical calculations, it is marked as a high-risk state and a resonance risk assessment report is generated for subsequent identification of unstable factors. Step S105: Based on the resonance risk assessment report generated in step S104, a support vector machine classification algorithm is used to classify and identify key factors that may cause instability in the system, extract high-risk parameter combinations related to the resonance phenomenon, and generate unstable factor distribution feature data for subsequent system simulation analysis; Step S106: Based on the unstable factor distribution characteristic data generated in step S105, a dynamic simulation model based on the Monte Carlo method is constructed to simulate the unstable evolution process under different parameter combinations in the operation of the power grid, obtain voltage and current fluctuation data in the simulated scenario, and calculate the system stability evaluation index for subsequent risk warning judgment; Step S107: Based on the system stability evaluation index calculated in step S106 and combined with historical operation data, determine whether the power grid operation is at an unstable boundary. If the evaluation index is lower than the safety threshold preset based on industry standards and historical experience, a risk warning signal is generated, and parameter adjustment suggestion data is provided.

[0005] Preferably, in step S101, acquiring multi-dimensional data from the power system through a multi-source sensor network and constructing an original data set containing time series features includes: Collect multi-dimensional data such as voltage, current, capacitance and inductance in real time from the power system to form an initial data set; For the initial data set, segmenting the data set according to the time window to generate multiple sub-data sets; Performing outlier detection on the sub-dataset using a preset threshold to determine the location of abnormal data points; If an abnormal data point is detected, the data in the time window where the abnormal data point is located is smoothed, and the mean filtering method is used to correct the data to obtain the smoothed time series data; Constructing a feature vector through the smoothed time series data to obtain a quantitative representation of the change trend of key parameters; Based on the quantitative representation of the change trend, a classification algorithm is applied to determine the potential system failure mode to obtain a classification result; If the classification results show the existence of a fault mode, the key features of the data in the relevant time window are extracted to determine the scope of the fault impact.

[0006] Preferably, in step S102, the preprocessing and denoising of the original data set, extracting the dynamic change characteristics of key parameters, and obtaining a cleaned parameter change sequence include: Performing standardization processing on the original data set using signal processing technology to generate a preliminarily sorted data set; Performing denoising on the preliminarily sorted data set by a filtering method to eliminate noise interference and obtain a denoised data set; Extracting characteristic information of capacitance and inductance values from the denoised data set, analyzing dynamic change trends based on time nodes, and generating a feature-extracted change sequence; Decomposing the parameter sequence of the feature-extracted change sequence using a time-frequency analysis method to obtain distribution characteristics of different frequency components; If there are abnormal frequency components in the distribution characteristics, the abnormal data are filtered out by a preset threshold to obtain filtered frequency distribution data; A time-frequency characteristic model is constructed based on the filtered frequency distribution data to generate a cleaned parameter change sequence.

[0007] Preferably, in step S103, analyzing the interaction relationship between key parameters based on the parameter change sequence after cleaning to obtain spectrum feature data includes: Obtaining parameter change data from the cleaned parameter change sequence, and generating a normalized parameter change set using a standardization processing method; For the normalized parameter change set, a time-frequency decomposition method is used to extract the frequency components of the capacitance and inductance values at different time scales to determine the time-frequency distribution characteristics; By using the time-frequency distribution characteristics, the frequency domain coupling relationship is calculated to obtain the correlation strength data between the two in each frequency band; Extracting intensity features from the correlation intensity data and performing quantization processing to generate a spectrum data set; If the intensity characteristics of certain frequency bands in the spectrum data set exceed a preset threshold, then mark the potential abnormal frequency band data; Based on the potential abnormal frequency band data, the correlation with the resonance risk is analyzed to determine the risk level distribution.

[0008] Preferably, in step S104, judging whether there is a resonance risk in the system based on the spectrum feature data and in combination with a resonance phenomenon theoretical model and generating a risk assessment report includes: Cleaning and normalizing the spectrum feature data to generate a processed spectrum feature data set; Based on the processed spectral feature data set and in combination with a pre-established theoretical model of resonance phenomenon, analyzing whether there are potential conditional features related to resonance; For the potential condition feature, calculating the deviation from a preset resonance threshold, and if the deviation exceeds a preset range, marking it as a data out-of-state; According to the data exceeding state, identifying possible unstable factors in the system, and using a classification algorithm to generate a classified set of unstable factors; If at least one factor in the unstable factor set meets the high-risk standard, the system state is marked as a high-risk state; Based on the high-risk state mark, a comprehensive risk assessment level and risk warning information are generated.

[0009] Preferably, in step S105, identifying unstable factors and extracting high-risk parameter combinations from the risk assessment report to generate unstable factor distribution feature data includes: Obtaining initial data from the risk assessment report, using a classification algorithm to preliminarily classify key factors that may cause instability, and generating a classified factor set; extracting high-risk parameter combinations associated with the resonance phenomenon based on the classified factor set, performing feature mapping processing, and generating a preliminary distribution feature data set; Cleaning and standardizing the preliminary distribution feature data set, and if the outliers exceed the preset threshold, smoothing is performed to generate the final distribution feature data; Performing correlation analysis on the final distribution feature data; if the correlation index is lower than a preset threshold, removing irrelevant parameters to obtain a streamlined parameter feature set; Based on the simplified parameter feature set, simulation input data is generated, multi-dimensional analysis is performed, and key parameter combinations are extracted.

[0010] Preferably, in step S106, constructing a dynamic simulation model based on the unstable factor distribution characteristic data, simulating fluctuations in power grid operation, and calculating system stability evaluation indicators includes: Based on the distribution characteristic data of the unstable factors, a dynamic simulation model is constructed using a random simulation method to generate initial simulation scene data and obtain the distribution information of the unstable evolution trajectory; Separating the voltage fluctuation information and the current fluctuation information from the distribution information, and determining independent feature sets for the two types of fluctuation information; Integrating the independent feature sets, and if the data integrity is lower than a preset threshold, performing interpolation processing to generate an integrated fluctuation feature data set; Based on the integrated fluctuation feature data set, a multi-dimensional analysis framework is constructed to obtain feature weights; If the feature weight distribution does not meet the preset balance condition, the weight is adjusted and the system stability evaluation index is recalculated; Screening and optimizing the outliers in the evaluation indicators to generate an optimized evaluation indicator data set.

[0011] Preferably, in step S107, judging whether the grid operation state is close to the instability boundary based on the system stability assessment index and generating a risk warning signal and parameter adjustment suggestions include: Obtain real-time system stability data from the power grid operation database, combine it with historical data for standardization, and generate a preliminary organized operating status data set; Calculate the system stability evaluation index based on the operating status data set, filter outliers, and determine the final evaluation index result; If the final evaluation index result is lower than the preset safety threshold, it is determined that the power grid operation is close to the unstable boundary, the risk warning mechanism is triggered, and a risk warning signal is generated; In response to the risk warning signal, by comparing and analyzing historical data and current operating status information, determine the key parameters that lead to the unstable state; Based on the key parameters, predict the parameter adjustment direction and generate parameter adjustment suggestion data; The optimized control strategy is calculated based on the parameter adjustment suggestion data and transmitted to the system control unit for parameter update.

[0012] The intelligent method for power security situation assessment based on a multi-source sensor network described in this application has the advantages of deploying a multi-source sensor network to collect key parameters of voltage, current, capacitance and inductance in real time, applying signal processing technology and wavelet transform to analyze the interaction between capacitance and inductance, and judging whether there are potential conditions for resonance in the system. The present invention further adopts a support vector machine algorithm to identify key factors that may cause instability, constructs a dynamic simulation model based on the Monte Carlo method to simulate the unstable evolution process, and calculates the system stability evaluation index. When the evaluation index is lower than the safety threshold, the present invention generates a risk warning signal and provides parameter adjustment suggestions, thereby realizing early warning of power system resonance risk and stability optimization control, and effectively improving the safety and reliability of power grid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is the process of an intelligent method for power security situation assessment based on a multi-source sensor network described in this application Figure 1 ; Figure 2This is the process of an intelligent method for power security situation assessment based on a multi-source sensor network described in this application Figure 2 . DETAILED DESCRIPTION

[0014] like Figure 1-Figure 2 As shown, the intelligent method for power security situation assessment based on a multi-source sensor network described in this application includes the following steps: like Figure 1-Figure 2 As shown, in step S101, by deploying a multi-source sensor network, multi-dimensional data is collected from the power system in real time, covering monitoring points of key parameters such as voltage, current, capacitance and inductance, and constructing a raw data set containing time series characteristics for subsequent dynamic change analysis.

[0015] Furthermore, in step S101, by deploying a multi-source sensor network, multi-dimensional data is collected from the power system in real time, covering monitoring points of key parameters such as voltage, current, capacitance, and inductance, and a raw data set containing time series characteristics is constructed to obtain a basic data set for subsequent analysis; Based on the constructed original data set, segmentation processing is performed based on time series characteristics, the data is divided into multiple sub-datasets according to the preset time window, and the time boundary of each sub-dataset is determined; Use the preset threshold method to detect outliers in each sub-dataset, analyze whether the data points are beyond the normal range, and obtain the location information of the abnormal data points; If an abnormal data point is detected, the data in the time window where the abnormal data point is located is smoothed, and the original data is corrected using the mean filtering method to obtain the smoothed time series data; By using smoothed time series data, the fluctuation of voltage and current parameters is analyzed, characteristic vectors are constructed, and a quantitative representation of the dynamic change trend is obtained; Based on the quantitative representation of dynamic change trends, the support vector machine algorithm is used to classify the data to determine whether there is a potential system failure mode and obtain the classification results; If the classification results indicate the existence of a fault mode, the multi-dimensional data within the relevant time window is deeply mined to extract key parameter features associated with the fault mode and determine the scope of the fault impact; By combining characteristic data of the fault impact range with the power system operation log, we can build association rules, analyze the dependencies between key parameters, and obtain a comprehensive assessment of the system operation status. According to the comprehensive evaluation results, signal processing technology is used to analyze the time-frequency characteristics of the dynamic change characteristics of the capacitance and inductance values, and the cleaned parameter change sequence is obtained.

[0016] Specifically, in step S101, a multi-source sensor network is deployed to collect multi-dimensional data of voltage, current, capacitance and inductance from the power system in real time with a sampling frequency of 1kHz, and a raw data set containing time series features is constructed to form a basic data set; Based on the original data set, the data is divided into multiple sub-datasets according to the time window. The length of each time window is 10 seconds, and the time boundary of each sub-dataset is determined; A preset threshold method is used to detect outliers in each sub-data set. The normal range of voltage is set to 220V±5%, and the normal range of current is set to 10A±2%. The data points are analyzed to see if they exceed the normal range, and the location information of the abnormal data points is obtained. If an abnormal data point is detected, the data in the time window where the abnormal data point is located is smoothed, and the original data is corrected using the mean filtering method with a window size of 5 data points to obtain the smoothed time series data; By analyzing the fluctuation of voltage and current parameters through smoothed time series data, the characteristic vector is constructed, and the mean and standard deviation are extracted as quantitative indicators to obtain a quantitative representation of the dynamic change trend; Based on the quantitative representation of dynamic change trends, the support vector machine algorithm is applied to classify the data, and the radial basis function kernel function is used to determine whether there is a potential system failure mode and obtain the classification result; If the classification results indicate a fault mode, the multi-dimensional data within the relevant time window is deeply mined to extract the mutation characteristics of the capacitance and inductance values and determine the scope of the fault impact; By combining characteristic data of the fault impact range with the power system operation log, we construct association rules, analyze the correlation coefficient between voltage and current, and obtain a comprehensive assessment of the system operation status. According to the comprehensive evaluation results, the fast Fourier transform is applied to perform time-frequency characteristic analysis on the dynamic change characteristics of capacitance and inductance values to obtain the parameter change sequence after cleaning.

[0017] In one embodiment, the specific implementation details of step S101 include: In the deployment of multi-source sensor networks, sensor types and configurations are as follows: Multi-source sensor networks are deployed at key nodes of the power system, including substations, distribution cabinets, and transmission lines. These include voltage sensors (accuracy ±0.2%), current sensors (accuracy ±0.5%), capacitance measurement modules (based on the LCR bridge principle), and inductance measurement modules (based on the resonant frequency method). Sampling frequency setting: Set the sampling frequency to 1kHz to meet the needs of capturing fast dynamic changes (harmonics, transient processes) in the power system; Data synchronization mechanism: GPS clock synchronization technology is used to ensure the timestamp consistency of multi-source sensor data with an error of ≤1ms; In data segmentation and anomaly detection, time window division: the original data is segmented into 10-second time windows, and each sub-dataset contains 10,000 data points (1kHz×10 seconds). Outlier detection: dynamic thresholds are set based on statistical methods: Normal voltage range: 220V±5% (i.e. 209V~231V); Normal current range: 10A±2% (i.e. 9.8A~10.2A); If a data point exceeds the threshold, it is marked as an anomaly and its time window position is recorded; In the mean filter processing of data smoothing and feature extraction, a sliding mean filter with a window size of 5 data points is used for abnormal window data. The formula is: , Character meaning: xi: original data point (instantaneous measurement value of voltage, current, capacitance or inductance), xsmooth: smoothed data point, k: index of the current data point, i: local index of the data point within the sliding window (range: k−2 to k+2); Feature vector construction: extract the mean, standard deviation, and peak value of voltage and current in each time window as dynamic change quantitative indicators to form a feature vector: , Character meaning: μV: mean voltage (reflects the average level of voltage), σV: standard deviation of voltage (reflects the degree of voltage fluctuation), Vpp: peak voltage (the difference between the maximum and minimum values, reflecting the dynamic range of voltage), μI: mean current, σI: standard deviation of current, Ipp: peak current; In the support vector machine (SVM) parameter settings for fault mode classification, the kernel function is radial basis function (RBF), the parameter γ is 0.1, the penalty coefficient C is 1.0, and the classification logic is: the training set contains historical normal data and fault data (short circuit, resonance), and the output classification results are "normal", "capacitance abnormality" or "inductance abnormality".

[0018] like Figure 1-Figure 2 As shown, in step S102, the original data set obtained in step S101 is preprocessed by applying signal processing technology, and the noise interference is removed by using filtering method, and the dynamic change characteristics of the capacitance value and the inductance value at different time points are extracted to obtain the cleaned parameter change sequence for subsequent time-frequency characteristic analysis.

[0019] Furthermore, in step S102, the original data set obtained from the initial stage is pre-processed using signal processing technology, and noise interference is removed from the data through filtering methods to obtain a preliminarily cleaned data sequence; Based on the data sequence after preliminary cleaning, the dynamic change characteristics of the capacitance and inductance values at different time points are extracted, and the parameter fluctuations are recorded using the time series analysis method to obtain the dynamic change feature set; By applying the normalization processing technology to the dynamic change feature set, the fluctuation data of capacitance and inductance values are normalized and adjusted to obtain the normalized parameter change sequence; According to the normalized parameter change sequence, the wavelet transform method is used to decompose the data into time and frequency components, extract the frequency components on different time scales, and determine the time-frequency distribution characteristic data; The distribution patterns of capacitance and inductance in each frequency band are analyzed through time-frequency distribution characteristic data. The frequency components are quantified using spectrum analysis tools to obtain a spectrum feature set. If the characteristic value of some frequency segments in the spectrum feature set exceeds the preset threshold range, the frequency segment is marked as abnormal and the potential abnormal frequency segment data is obtained; Based on the potential abnormal frequency band data, combined with the pre-built risk assessment model, the abnormal frequency bands are analyzed for correlation and the resonance risk level distribution is determined; Through the distribution of resonance risk levels, the corresponding spectrum feature mapping relationship is generated, and the mapping relationship is structured using data visualization technology to obtain risk assessment basis data; Based on the risk assessment data, the cleaned parameter change sequence is verified in combination with the time-frequency characteristic description model to determine whether the final data characteristics meet the predetermined standards.

[0020] Specifically, in step S102, a Butterworth low-pass filter is used to pre-process the original data set, with a cutoff frequency of 100 Hz to remove high-frequency noise interference, thereby obtaining a preliminary cleaned data sequence with a signal-to-noise ratio improved by 20 dB. Based on the cleaned data sequence, the sliding window method is used to extract the dynamic change characteristics of the capacitance and inductance values. The window width is set to 50ms and the sampling interval is 10ms. The parameter fluctuation curve is recorded to form a dynamic feature set containing 1000 time points. Through the dynamic feature set, the Z-score normalization algorithm is applied to normalize the capacitance and inductance fluctuation data, the mean is adjusted to 0 and the standard deviation is 1, and a normalized parameter sequence is generated; Based on the normalized sequence, the Daubechies wavelet basis function is used to perform a five-layer time-frequency decomposition, extracting the frequency components of the three frequency bands of 0-50Hz, 50-100Hz, and 100-200Hz. The energy proportion of each scale is calculated to form a time-frequency distribution matrix. Through the time-frequency matrix, the fast Fourier transform is used to analyze the amplitude correlation of capacitance and inductance in the 60Hz and 120Hz frequency bands, and the frequency bands with Pearson correlation coefficient greater than 0.8 are calculated to output the spectrum feature set. If the amplitude of the 120Hz frequency band in the spectrum set exceeds the threshold of 0.5V, it is marked as an abnormal frequency band, and the segment lasting more than 200ms is recorded as potential abnormal data; Based on the abnormal data, the pre-trained random forest risk assessment model is called, and the three features of frequency band amplitude, duration, and correlation coefficient are input to output the distribution of frequency bands with a high risk probability greater than 80%; Through risk distribution, a two-dimensional mapping table of frequency and risk level is constructed, and a heat map algorithm is used to generate a visualization matrix, with the horizontal axis representing the 50-200 Hz frequency band and the vertical axis representing the risk level 1-5; Based on the visualization matrix and combined with the frequency domain stability index of the time-frequency characteristic model, when the standard deviation of the fluctuation in the 60Hz frequency band is less than 0.1 and the risk level in the 120Hz frequency band is ≤3, the data is judged to comply with the IEC 61000-4-30 standard.

[0021] In one embodiment, the specific implementation details of step S102 include: In filtering and denoising, Butterworth low-pass filter: Design a 4th-order filter with a cutoff frequency of 100Hz, a passband ripple ≤ 0.1dB, and a stopband attenuation ≥ 40dB. Signal-to-noise ratio improvement: Through filtering, the signal-to-noise ratio (SNR) is improved by 20dB, effectively suppressing high-frequency noise (switching noise, electromagnetic interference). In dynamic feature extraction, the sliding window method sets the window width to 50ms and the sampling interval to 10ms to extract the instantaneous rate of change of capacitance and inductance values: , Symbol description: C: capacitance value (unit: Farad, F), L: inductance value (unit: Henry, H), Δt: sampling time interval (unit: second, s), , : The instantaneous rate of change of capacitance and inductance values (reflecting the dynamic characteristics of the parameters); In data standardization, Z-score standardization is used: , Symbol explanation: μC: mean capacitance value, σC: standard deviation of capacitance value, μL: mean inductance value, σL: standard deviation of inductance value, C norm 、Lnorm : Normalized capacitance and inductance values (dimensionless); In the wavelet time-frequency decomposition, the wavelet basis function is Daubechies wavelet (db4), the decomposition layer is 5, and the frequency band is covered from 0.1Hz to 10kHz. The energy proportion is calculated by using the wavelet coefficients to calculate the energy proportion of each frequency band. The formula is: , Explanation of symbols: Wk(i): coefficients after wavelet decomposition at the kth layer (reflecting the energy distribution of the signal in a specific frequency band), Ek: total energy of the wavelet coefficients at the kth layer (used to quantify the importance of the frequency band), N: number of wavelet coefficients.

[0022] like Figure 1-Figure 2 As shown, step S103, according to the parameter change sequence after cleaning obtained in step S102, the interaction relationship between the capacitance value and the inductance value is analyzed using a time-frequency analysis model based on wavelet transform, the coupling characteristics of the two in the frequency domain are calculated, and the spectrum characteristic data reflecting the interaction strength is obtained for subsequent resonance risk assessment.

[0023] Furthermore, in step S103, according to the cleaned parameter change sequence, a time-frequency analysis model based on wavelet transform is used to decompose the time series data of the capacitance and inductance values to obtain the frequency component distribution of the two at different time scales; By analyzing the dynamic variation characteristics of capacitance and inductance at various time scales through frequency component distribution, the distribution patterns of the two in the time-frequency domain are determined. According to the distribution law in the time-frequency domain, the coupling characteristics of the capacitance and inductance in the frequency domain are calculated to obtain the spectrum characteristic data reflecting the interaction strength; Through the spectrum characteristic data, the spectrum analysis tool is used to quantify the coupling characteristics and obtain the correlation strength value of each frequency band; If the correlation strength value exceeds the preset threshold in certain frequency bands, the frequency band will be marked as abnormal to determine the potential high-risk frequency band data; Based on the high-risk frequency band data and the pre-established resonance risk assessment model, the correlation between the spectrum characteristics and the resonance phenomenon is analyzed to obtain the risk level distribution; Through the risk level distribution, the corresponding spectrum feature mapping relationship is generated to determine the key frequency band characteristics related to the resonance risk; Based on the characteristics of key frequency bands and combined with dynamic simulation data of system operating parameters, the frequency band characteristics are verified in multiple dimensions to obtain the verified characteristic distribution results; Through the verified characteristic distribution results, the parameter combination directly related to the resonance risk is extracted to generate the final risk assessment basis data.

[0024] Specifically, in step S103, according to the parameter change sequence after cleaning, the time-frequency analysis model based on wavelet transform is used to decompose the time series data of capacitance and inductance values, and the Daubechies wavelet basis function is used to perform multi-scale decomposition on the data to obtain the frequency component distribution of the two in the range of 0.1 Hz to 10 kHz; The dynamic variation characteristics of capacitance and inductance at various time scales are analyzed through frequency component distribution. The energy distribution in the time-frequency domain is calculated using short-time Fourier transform (STFT) to determine the distribution patterns of the two in the time-frequency domain. Based on the distribution patterns in the time-frequency domain, the coupling characteristics of the capacitance and inductance in the frequency domain are calculated. The cross-spectral density function is used to analyze the interaction strength between the two in the frequency range of 1kHz to 5kHz, and the spectral characteristic data reflecting the interaction strength is obtained. The coupling characteristics are quantified using the fast Fourier transform (FFT) through the spectral feature data to obtain the correlation strength value of each frequency band. For example, the correlation strength in the 2kHz band is 0.85. If the correlation strength value exceeds the preset threshold (such as 0.8) in certain frequency bands, the frequency band will be marked as abnormal to identify potential high-risk frequency band data; Based on the high-risk frequency band data and the pre-established resonance risk assessment model, the support vector machine (SVM) algorithm is used to analyze the correlation between spectrum characteristics and resonance phenomena to obtain the risk level distribution. For example, the risk level in the 3kHz frequency band is high. Through the risk level distribution, the corresponding spectrum feature mapping relationship is generated to determine the key frequency band characteristics related to the resonance risk. For example, the characteristic value in the 4kHz frequency band is 0.92; Based on the characteristics of key frequency bands and combined with dynamic simulation data of system operating parameters, the MATLAB simulation tool is used to conduct multi-dimensional verification of the frequency band characteristics and obtain the verified characteristic distribution results. For example, the verification result in the 5kHz frequency band is 0.89; Through the verified characteristic distribution results, the parameter combinations directly related to the resonance risk are extracted to generate the final risk assessment basis data. For example, the parameter combination in the 6kHz frequency band is a capacitance value of 10μF and an inductance value of 50mH.

[0025] In one embodiment, the specific implementation details of step S103 include: In the cross-spectral density calculation, the cross-spectral density (CSD) formula of capacitance and inductance is: , Explanation of symbols: R CL(τ): Cross-correlation function of capacitance C and inductance L (reflecting the time domain correlation between the two), F: Fourier transform (converting time domain signals into frequency domain signals), S CL (f): Cross-spectral density of capacitance and inductance at frequency f (quantifying the coupling strength in the frequency domain); In the correlation strength quantification, the Pearson correlation coefficient is calculated in the 1kHz~5kHz frequency band: , Explanation of symbols: Cov(C(f), L(f)): covariance of capacitance and inductance at frequency f (reflecting linear correlation), σC, σL: standard deviation of capacitance and inductance values, ρ: correlation coefficient (range: -1 to 1, larger absolute values indicate stronger correlation); In the high-risk frequency band marking, the preset threshold is: the frequency band with correlation strength ≥ 0.8 or amplitude ≥ 0.5V is marked as abnormal. For example, the 2kHz frequency band with correlation strength 0.85 is marked as high risk.

[0026] like Figure 1-Figure 2 As shown, in step S104, based on the spectrum feature data obtained in step S103, combined with the theoretical model of the resonance phenomenon, it is determined whether there are potential conditions for resonance to occur in the system. If the spectrum feature data exceeds the resonance threshold preset based on historical data and theoretical calculations, it is marked as a high-risk state, and a resonance risk assessment report is generated for subsequent identification of unstable factors.

[0027] Furthermore, in step S104, spectral feature data is extracted from the system, and the data is cleaned and normalized using a standardization processing method to obtain a standardized spectral feature data set; Based on the normalized spectral feature dataset and the pre-built theoretical model of the resonance phenomenon, we analyze whether there are potential conditional features related to resonance in the data and determine the distribution of the potential conditions. Based on the determined potential condition characteristics, combined with theoretical calculation results based on historical data, the deviation between the spectrum feature data set and the preset resonance threshold is calculated to obtain a deviation data set; If the deviation value in the deviation data set exceeds the preset threshold range, it is marked as a data out-of-state and a preliminary abnormal state record is generated; Based on the records of data exceeding the state, combined with the system analysis method, the possible unstable factors in the system are extracted to obtain a preliminary set of unstable factors; For the preliminary set of unstable factors, support vector machine algorithm is used to classify the factors and determine the set of unstable factors after classification; If at least one factor in the classified unstable factor set meets the high-risk state standard, the system state is marked as a high-risk state and a high-risk state identifier is generated; Based on the high-risk status identification and combined with the resonance risk assessment logic, a comprehensive analysis of the potential resonance phenomenon in the system is conducted to obtain the final risk assessment level; Through the final risk assessment level, a structured resonance risk assessment report is generated, and the instability factors and potential condition data are associated to output a complete risk analysis data set.

[0028] Specifically, in step S104, spectral feature data is extracted from the system, and the data is cleaned and normalized using a standardization processing method, for example, the frequency range is limited to 0 Hz to 100 Hz, and the Z-score method is used to convert the data into a normalized spectral feature dataset with a mean of 0 and a standard deviation of 1; Based on the normalized spectral feature dataset and pre-built theoretical models of resonance phenomena, such as the RLC circuit model, the data is analyzed to determine whether there are potential conditional features related to resonance and the distribution of potential conditions, such as a frequency peak near 50 Hz. For the determined potential conditional features, combined with theoretical calculation results based on historical data, for example, calculating the deviation between the spectrum feature dataset and a preset resonance threshold (such as an amplitude threshold of 0.8), a deviation data set is obtained; If the deviation value in the deviation data set exceeds the preset threshold range, for example, the amplitude deviation of a certain frequency point reaches 1.2, it is marked as a data exceedance state and a preliminary abnormal state record is generated; Based on the records of data exceeding the state, combined with system analysis methods such as principal component analysis (PCA), possible unstable factors in the system are extracted to obtain a preliminary set of unstable factors, such as frequency fluctuations and amplitude anomalies; For the preliminary set of unstable factors, the support vector machine algorithm is used to classify the factors, for example, using the radial basis function (RBF) kernel function for classification, to determine the set of classified unstable factors, such as high-frequency fluctuations and low-frequency resonances; If at least one factor in the classified unstable factor set meets the high-risk state standard, for example, the risk score of high-frequency fluctuation exceeds 90 points, the system state is marked as high-risk and a high-risk state indicator is generated; Based on the high-risk status identification, combined with the resonance risk assessment logic, such as the weighted scoring method, a comprehensive analysis of the potential resonance phenomenon in the system is performed to obtain the final risk assessment level, such as the high risk level of 5; Based on the final risk assessment level, a structured resonance risk assessment report is generated, and unstable factors and potential condition data are associated, such as associating high-frequency fluctuations with the 50Hz frequency point, to output a complete risk analysis data set.

[0029] In one embodiment, the specific implementation details of step S104 include: In the RCL series resonance model, the resonant frequency calculation formula is: , Explanation of symbols: L: inductance (H), C: capacitance (F), fr: resonant frequency (Hz), i.e. the critical frequency at which the system may resonate; In principal component analysis (PCA), the spectral feature data is reduced in dimension, the principal components with a contribution rate greater than 85% are retained, and key unstable factors (such as frequency offset and harmonic distortion) are extracted. In the risk level classification, support vector machine (SVM) classification is used, and the risk level is divided into 5 levels: level 1 (safe) to level 5 (emergency).

[0030] like Figure 1-Figure 2 As shown, in step S105, based on the resonance risk assessment report generated in step S104, a support vector machine classification algorithm is used to classify and identify key factors that may cause instability in the system, extract high-risk parameter combinations related to the resonance phenomenon, and generate unstable factor distribution feature data for subsequent system simulation analysis.

[0031] Furthermore, in step S105, based on the resonance risk assessment report generated in step S104, the system operating parameters and risk indicator data recorded therein are extracted, and a support vector machine classification algorithm is used to classify and identify key factors that may cause instability, thereby obtaining a classified set of instability factors; By combining the classified unstable factor set with the theoretical model of the resonance phenomenon, we analyze the correlation between each factor and the resonance risk, extract the high-risk parameter combinations related to the resonance phenomenon, and determine the high-risk parameter combination data set; Based on the high-risk parameter combination data set, mapping processing is performed on the parameter characteristics therein to generate preliminary unstable factor distribution characteristic data and obtain a preliminary distribution characteristic data set; The preliminary distribution feature data set is used for data cleaning and standardization to remove noise data and outliers. If the outlier ratio exceeds the preset threshold, the outliers are smoothed to obtain the cleaned distribution feature data set. Through the cleaned distribution feature data set, correlation analysis is performed on high-risk parameter combinations. If the correlation index is lower than the preset threshold, irrelevant parameters are eliminated to determine the streamlined parameter feature set; Based on the simplified parameter feature set, the unstable factors are dynamically mapped in combination with the system simulation environment to generate mapped simulation input data and judge the integrity and consistency of the data; Using the mapped simulation input data and combining it with the pre-established simulation model, we conduct a multi-dimensional analysis of the distribution characteristics of unstable factors and obtain the characteristic distribution results after simulation analysis. Through the characteristic distribution results after simulation analysis, the key parameter combinations directly related to resonance risk are extracted, the final unstable factor distribution characteristic data is generated, and the potential risk points in system operation are determined; Based on the final distribution characteristic data of unstable factors, combined with risk assessment logic and historical data analysis results, structured risk analysis data is generated to obtain the input data set for subsequent system optimization.

[0032] Specifically, in step S105, system operating parameters and risk indicator data are extracted from the resonance risk assessment report, including voltage fluctuation range (e.g., ±5%), frequency deviation (e.g., ±0.2Hz), and harmonic distortion rate (e.g., THD>3%). A support vector machine classification algorithm (RBF kernel function is selected, with penalty coefficient C=1.0) is used to classify key factors and identify a set of unstable factors such as capacitor bank switching and nonlinear loads. Combined with resonance theory models (such as the RLC series resonance formula), calculate the correlation coefficient between each factor and resonance risk (such as the Pearson coefficient > 0.7), and screen out high-risk parameter combinations (such as a combination of capacitance C > 100μF and inductance L < 10mH); Perform feature mapping on the parameter combination (such as principal component analysis (PCA) to reduce the dimension to 3 dimensions) to generate preliminary distribution feature data including statistical quantities such as mean and variance; The Z-score standardization method was used to clean the data. If the outlier ratio exceeded 5%, the sliding window average method was used for smoothing to obtain a cleaned data set with a standard deviation less than 1. Calculate the support between parameters through correlation analysis (such as the Apriori algorithm). If the support is less than 0.3, remove irrelevant parameters (such as temperature parameters) and retain a simplified feature set (such as voltage, frequency, and harmonic content). Input the feature set into a simulation environment (such as MATLAB / Simulink), dynamically map it to simulation input data (such as time domain waveform), and check the data integrity (missing rate <1%) and consistency (correlation coefficient >0.9); Perform multi-dimensional analysis (such as frequency domain FFT and time domain step response) based on simulation models (such as state-space models) and output distribution results such as resonant peak frequency (such as 250 Hz) and amplitude (such as 120 dB). Extract key parameter combinations (such as capacitance C = 150 μF and harmonic order h = 5), generate final distribution characteristic data (such as Gaussian distribution μ = 230 Hz, σ = 15 Hz), and mark potential risk points (such as the resonant point frequency close to the system's natural frequency); Combine historical data (such as the results of the past 100 simulations) and risk assessment logic (such as a risk matrix) to output structured data (such as JSON format), including risk levels (such as high risk), parameter thresholds, and optimization suggestions.

[0033] In one embodiment, the specific implementation details of step S105 include: In parameter combination screening, an example of a high-risk combination is: capacitance value C>100μF and inductance value L<10mH, verified by the Pearson correlation coefficient (>0.7); In data dimensionality reduction and mapping, principal component analysis (PCA) reduces features to 3 dimensions and retains components that account for > 90% of the variance; In simulation input generation, a state space model is constructed in MATLAB / Simulink, and the input parameters include voltage, frequency, and harmonic content (THD).

[0034] like Figure 1-Figure 2 As shown, in step S106, for the unstable factor distribution characteristic data generated in step S105, a dynamic simulation model based on the Monte Carlo method is constructed to simulate the unstable evolution process under different parameter combinations in the operation of the power grid, obtain the voltage and current fluctuation data in the simulation scenario, and calculate the system stability evaluation index for subsequent risk warning judgment.

[0035] Furthermore, in step S106, the distribution characteristic data of the unstable factors is obtained from step S105, and a dynamic simulation model is constructed using the Monte Carlo method to simulate the unstable evolution process under different parameter combinations in the operation of the power grid, thereby obtaining preliminary voltage and current fluctuation data in the simulation scenario; Based on the obtained preliminary voltage and current fluctuation data, data standardization is performed, and calibration is performed in combination with historical operating status information to determine the distribution characteristics of the calibrated fluctuation data; The system stability evaluation index is calculated based on the distribution characteristics of the calibrated fluctuation data, and the index value is quantified using a preset evaluation algorithm to obtain the quantified stability evaluation result; If the quantified stability assessment result is lower than the preset safety threshold, the data in-depth analysis mechanism is triggered to obtain historical unstable event data related to the current simulation scenario and determine the potential risk level; Based on the historical instability event data obtained and the fluctuation data distribution characteristics of the current simulation scenario, cluster analysis methods are used to classify risk factors and determine the main instability influencing parameters; By determining the main instability influencing parameters, a parameter sensitivity analysis model is constructed to simulate the impact of different parameter adjustments on system stability and obtain preliminary impact data of parameter adjustments; Based on the obtained preliminary impact data of parameter adjustment and combined with real-time grid operation status information, an optimization algorithm is used to calculate the control strategy suitable for the current scenario and determine the optimized parameter adjustment plan; Through the determined optimized parameter adjustment plan, the corresponding risk warning signal is generated and synchronously transmitted to the system control module to obtain the updated operating status data after feedback; Based on the acquired operating status update data, the execution results of the adjustment plan are compared and analyzed with the historical database, and the machine learning algorithm is used to update the distribution characteristics of unstable factors to obtain the improved risk prediction model data.

[0036] Specifically, in step S106, the distribution characteristic data of the unstable factors is obtained from step S105, and a dynamic simulation model is constructed using the Monte Carlo method to simulate the unstable evolution process under different parameter combinations in the operation of the power grid. For example, under the conditions of a voltage fluctuation range of ±5% and a current fluctuation range of ±3%, 1000 random simulation scenarios are generated to obtain preliminary voltage and current fluctuation data; Based on the preliminary voltage and current fluctuation data obtained, data standardization is performed and calibration is performed in combination with historical operating status information. For example, the Z-score standardization method is used to convert the fluctuation data into a distribution with a mean of 0 and a standard deviation of 1, and the distribution characteristics of the calibrated fluctuation data are determined; The system stability evaluation index is calculated based on the calibrated fluctuation data distribution characteristics, and the index value is quantified using a preset evaluation algorithm. For example, the Lyapunov exponent method is used to calculate the system stability to obtain a quantified stability evaluation result. If the quantified stability assessment result is lower than the preset safety threshold, for example, the Lyapunov exponent is less than -0.5, the data in-depth analysis mechanism is triggered to obtain historical instability event data related to the current simulation scenario, such as 10 voltage collapse events that occurred in the past year, to determine the potential risk level; Based on the historical instability event data obtained and the fluctuation data distribution characteristics of the current simulation scenario, cluster analysis methods are used to classify risk factors. For example, the K-means algorithm is used to classify risk factors into three categories and determine the main instability influencing parameters; By identifying the main instability-influencing parameters, a parameter sensitivity analysis model was constructed to simulate the impact of different parameter adjustments on system stability. For example, adjusting the voltage fluctuation range to ±3% and the current fluctuation range to ±2% provided preliminary impact data for parameter adjustments. Based on the obtained preliminary impact data of parameter adjustments and combined with real-time grid operation status information, an optimization algorithm is used to calculate a control strategy suitable for the current scenario. For example, a genetic algorithm is used to optimize control parameters and determine the optimized parameter adjustment plan. Through the determined optimized parameter adjustment plan, the corresponding risk warning signal is generated and synchronously transmitted to the system control module to obtain the updated operating status data after feedback; Based on the acquired operating status update data, the execution results of the adjustment plan are compared and analyzed with the historical database, and the distribution characteristics of unstable factors are updated using a machine learning algorithm. For example, a new risk prediction model is trained using a random forest algorithm to obtain improved risk prediction model data.

[0037] In one embodiment, the specific implementation details of step S106 include: In the random scenario generation, Monte Carlo simulations were performed 1000 times, and the parameter fluctuation range was: voltage ±5%, current ±3%; Among the stability evaluation indicators, the Lyapunov exponent is calculated using the following formula: , Explanation of symbols: δx(t): small initial deviation of the system state (e.g., small change in voltage or current), λ: Lyapunov exponent (reflecting the system's sensitivity to initial conditions): λ<0: system is stable, λ>0: system is chaotic or unstable, and the safety threshold is set to λ<-0.5; In the parameter sensitivity analysis, a genetic algorithm is used to optimize the control parameters, and the objective function is to minimize the voltage fluctuation.

[0038] like Figure 1-Figure 2 As shown, step S107, based on the system stability evaluation index calculated in step S106 and combined with historical operation data, determines whether the power grid operation is at an unstable boundary. If the evaluation index is lower than the safety threshold preset based on industry standards and historical experience, a risk warning signal is generated, and parameter adjustment suggestion data is provided for subsequent system control optimization.

[0039] Furthermore, in step S107, real-time system stability related data is extracted from the power grid operation database, and preliminary cleaning and standardization processing is performed in combination with historical operation records to obtain a sorted operation status data set; Based on the collated operating status data set, the pre-built evaluation model is applied to perform calculations, outliers in the calculation process are filtered out, and preliminary results of the system stability evaluation indicators are determined; If the preliminary results of the system stability assessment indicators are lower than the safety thresholds preset based on industry standards and historical experience, the grid operation is judged to be close to the unstable boundary, triggering the risk warning mechanism and generating an initial risk warning signal; Through the initial risk warning signal, obtain historical operating data and current status information related to the unstable boundary, conduct comparative analysis, and determine the key parameters that may lead to unstable state; Based on the determined key parameters, the support vector machine algorithm is used to predict the parameter adjustment direction and generate preliminary parameter adjustment recommendation data; Based on the preliminary parameter adjustment recommendations and real-time feedback from the system control module, a dynamic simulation model based on the Monte Carlo method was constructed to simulate the unstable evolution process under different parameter combinations and obtain the voltage and current fluctuation data in the simulation scenario. By simulating voltage and current fluctuation data in the scenario, the system stability assessment index is recalculated and combined with historical operation data for comprehensive analysis to determine whether the grid operation is at the unstable boundary. If the recalculated system stability assessment index is lower than the preset safety threshold, a final risk warning signal is generated, and the parameter adjustment recommendation data is updated at the same time for reference in subsequent system control optimization; The final risk warning signal and updated parameter adjustment recommendation data are transmitted to the system control unit for parameter update, and the adjusted operating status information is simultaneously recorded and stored in the historical database for subsequent analysis.

[0040] Specifically, in step S107, real-time voltage, frequency, and power data are extracted from the power grid operation database, combined with 10-minute sampling data in the historical operation records, and processed using the Z-score normalization method to eliminate dimensional differences and form a standardized data set; The standardized data is input into the pre-trained LSTM evaluation model to calculate the system stability index. The safety threshold is set to 0.85. If the index falls below the threshold, an alert is triggered. Using early warning signals, retrieve similar abnormal data from the past 30 days and use Pearson correlation coefficient analysis to identify key parameters such as bus voltage deviation exceeding ±5% or frequency fluctuation greater than 0.2Hz; Based on key parameters, the SVM algorithm is used, with the kernel function set to RBF and the penalty coefficient C = 1.0. Output parameter adjustment suggestions include increasing generator output by 3% or switching capacitor banks. In combination with real-time SCADA feedback, a Monte Carlo model was constructed to simulate 1,000 parameter combinations, recording abnormal scenarios where the voltage dropped below 0.9 pu or the current distortion rate exceeded 5%. The transient stability margin index was recalculated using simulation data. When the index value remained below 0.82 for 2 seconds, a final warning was generated, and an optimization adjustment was recommended to correct the generator adjustment coefficient by 0.05. The adjustment instructions are sent to the PMU device through the IEC 61850 protocol, and the adjusted data is synchronously stored in the Oracle database and marked as an unstable boundary case.

[0041] In one embodiment, the specific implementation details of step S107 include: In the generation of early warning signals, real-time data is compared with the historical database (Oracle). If the Lyapunov exponent is less than -0.5, an early warning is triggered; In the parameter adjustment suggestions: Example: generator output increased by 3%, capacitor bank switching plan optimized; When issuing control instructions, the adjustment instructions are transmitted to the PMU (synchronized phasor measurement unit) through the IEC 61850 protocol.

[0042] Those skilled in the art can make various other corresponding changes and deformations based on the technical solutions and concepts described above, and all of these changes and deformations should fall within the scope of protection of the claims of this application.

Claims

1. An intelligent method for power security situation assessment based on multi-source sensor network, characterized in that: include: Acquire multi-dimensional data from the power system through a multi-source sensor network. The multi-dimensional data includes voltage, current, capacitance, and inductance, and construct a raw data set containing time series features. Preprocessing and denoising the original data set, extracting dynamic change characteristics of capacitance and inductance values, and obtaining a cleaned parameter change sequence; Analyzing the interaction between capacitance and inductance based on the parameter change sequence after cleaning to obtain spectrum characteristic data; Based on the spectrum characteristic data and in combination with a theoretical model of the resonance phenomenon, determine whether there is a resonance risk in the system and generate a risk assessment report; Based on the risk assessment report, identifying unstable factors and extracting high-risk parameter combinations to generate unstable factor distribution characteristic data; A dynamic simulation model is constructed based on the distribution characteristic data of the unstable factors to simulate the voltage and current fluctuations in the operation of the power grid and calculate the system stability evaluation index; Based on the system stability assessment indicators, it is determined whether the power grid operation state is close to the unstable boundary, and risk warning signals and parameter adjustment suggestions are generated.

2. According to claim 1, an intelligent method for power security situation assessment based on a multi-source sensor network is characterized in that: The deployment of the multi-source sensor network includes: Deploy voltage sensors, current sensors, capacitance measurement modules, and inductance measurement modules at substations, distribution cabinets, and key nodes of transmission lines in the power system; The accuracy of the voltage sensor is ±0.2%, the accuracy of the current sensor is ±0.5%, the capacitance measurement module is based on the LCR bridge principle, and the inductance measurement module is based on the resonant frequency method; The sampling frequency of the multi-source sensor network is set to 1kHz, and the GPS clock synchronization technology is used to ensure the consistency of data timestamps with an error of ≤1ms; The original data set is divided into multiple sub-data sets according to 10-second time windows, and each sub-data set contains 10,000 data points; The dynamic thresholds for abnormal value detection are: normal voltage range 220V±5%, normal current range 10A±2%; The abnormal window data is processed by sliding mean filtering with a window size of 5 data points; The mean, standard deviation and peak value of voltage and current in each time window are extracted to construct the feature vector; The data were classified using the support vector machine algorithm. The kernel function was the radial basis function (RBF), the parameter γ = 0.1, and the penalty coefficient C = 1.

0. The classification results included normal, abnormal capacitance, or abnormal inductance.

3. According to claim 1, an intelligent method for power security situation assessment based on a multi-source sensor network is characterized in that: The construction and stability evaluation of the dynamic simulation model include: The Monte Carlo method was used to generate 1,000 random simulation scenarios, simulating grid operation under conditions of voltage fluctuations within ±5% and current fluctuations within ±3%. The system stability evaluation index was calculated using the Lyapunov exponent method. When the Lyapunov exponent is less than -0.5, the in-depth analysis mechanism is triggered. Combined with historical unstable event data, the K-means clustering algorithm is used to classify risk factors into three categories; A parameter sensitivity analysis model was constructed, and a genetic algorithm was used to optimize the control parameters. The objective function was to minimize voltage fluctuations. The optimized parameter adjustment scheme was transmitted to the synchronized phasor measurement unit via the IEC61850 protocol, and the adjusted data was recorded in a historical database.

4. The intelligent method for power security situation assessment based on a multi-source sensor network according to claim 1 is characterized in that: The preprocessing and denoising include: The original data set was denoised using a 4th-order Butterworth low-pass filter with a cutoff frequency of 100 Hz, a passband ripple ≤ 0.1 dB, and a stopband attenuation ≥ 40 dB; The instantaneous rate of change of capacitance and inductance values is extracted by sliding window method, with a window width of 50ms and a sampling interval of 10ms; The Z-score normalization method is used to normalize the capacitance and inductance values.

5. The intelligent method for power security situation assessment based on a multi-source sensor network according to claim 1 is characterized in that: The acquisition of the spectrum characteristic data includes: The Daubechies wavelet basis function db4 is used to perform a 5-layer time-frequency decomposition of the parameter change sequence, covering the frequency band from 0.1Hz to 10kHz; The interaction strength between capacitors and inductors in the 1kHz to 5kHz frequency range is analyzed using the cross-spectral density function. If the frequency band correlation strength is ≥0.8 or the amplitude is ≥0.5V, it is marked as an abnormal frequency band; The support vector machine algorithm is used to analyze the correlation between spectrum characteristics and resonance risks. The risk level is divided into 5 levels, and a high-risk warning is generated when the risk level is ≥ 4.

6. The intelligent method for power security situation assessment based on a multi-source sensor network according to claim 1 is characterized in that: The risk assessment report generation includes: Calculate the resonant frequency based on the RLC series resonance model; The spectral feature data is reduced in dimension by principal component analysis (PCA), and the principal components with a variance contribution rate greater than 85% are retained; When the high-frequency fluctuation risk score is greater than 90 points, the system status is marked as high-risk and a comprehensive report including frequency deviation and harmonic distortion rate is generated.

7. The intelligent method for power security situation assessment based on a multi-source sensor network according to claim 1 is characterized in that: Generating the parameter adjustment suggestion includes: The long short-term memory network (LSTM) model is used to calculate the real-time system stability index, and the safety threshold is set to 0.85; When the bus voltage deviation exceeds ±5% or the frequency fluctuation is greater than 0.2Hz, the generator output or capacitor bank switching plan is optimized through genetic algorithms; The adjusted parameters are sent to the control unit via the IEC 61850 protocol and marked as unstable boundary cases in the Oracle database.

8. The intelligent method for power security situation assessment based on a multi-source sensor network according to claim 2 is characterized in that: The training set of the support vector machine algorithm includes historical normal data and fault data, and the fault types include short circuit and resonance; When the classification result is normal, capacitance abnormal, or inductance abnormal, deep data mining of the corresponding time window is triggered.

9. The intelligent method for power security situation assessment based on a multi-source sensor network according to claim 3 is characterized in that: The number of iterations of the genetic algorithm was set to 1000, the crossover probability was 0.8, and the mutation probability was 0.01; the optimized parameter combination included adjusting the voltage fluctuation range to ±3% and the current fluctuation range to ±2%.

10. The intelligent method for power security situation assessment based on a multi-source sensor network according to claim 5, characterized in that: When the duration of the high-risk frequency band is greater than 200ms, the risk probability is calculated. If the probability is greater than 80%, an emergency control instruction is generated and synchronized to the power grid dispatching center.

Images