A method for detecting the operating state of a low-voltage distribution network based on a fusion terminal

Abstract

The present invention discloses a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal, which relates to the technical field of distribution networks. The method includes: the fusion terminal acquires the original data set of the low-voltage distribution network; performs data preprocessing on the original data set; extracts features from the preprocessed data; performs feature fusion on the extracted features to obtain a fused feature vector, and divides the training set and the test set; constructs a neural network classification model for training to obtain a state classification model, and optimizes the state classification model, inputs the fused feature vector into the state classification model for classification; selects abnormal state data for detection to identify the types of abnormal states; selects normal state data for health assessment, and comprehensively evaluates the health state of the low-voltage distribution network through health key indicators and a health assessment function. The present invention improves the monitoring accuracy of the low-voltage distribution network, enhances the abnormal detection and health assessment capabilities, and optimizes the resource allocation.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution networks, and particularly to a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal. Background Art

[0002] The current detection of the operating state of low-voltage distribution networks relies on a number of key technologies. Among them, smart meters have become an indispensable part of low-voltage distribution networks, and through data collection, communication, and monitoring functions, user electricity consumption information and the operating state of distribution equipment are obtained. Sensor technology is widely used in the detection of electrical energy parameters such as current, voltage, power factor, and physical quantities such as equipment temperature and humidity. Data communication technology realizes remote data transmission and centralized management. Data analysis and processing technology extracts useful information and judges the operating state through methods such as statistical analysis, pattern recognition, and machine learning. Cloud computing and big data technology are applied to data storage, processing, and analysis to realize the centralized management, real-time monitoring, and remote control of distribution network data. The comprehensive application of these technologies improves the safety, stability, and operating efficiency of low-voltage distribution networks. Summary of the Invention

[0003] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further described in detail in the Detailed Description section. The Summary of the Invention section of the present invention does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the protection scope of the claimed technical solution.

[0004] Due to the lack of effective real-time monitoring means in traditional low-voltage distribution networks, it is difficult to locate and handle faults in a timely and accurate manner after they occur. Existing distribution network monitoring methods may not be able to comprehensively reflect the operating conditions of the distribution network, resulting in difficulties in precisely carrying out maintenance work. The present invention proposes a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal, and the above method includes:

[0005] S10: The fusion terminal obtains the original data set of the low-voltage distribution network;

[0006] S20: Perform data preprocessing on the original data set to obtain a denoised and filled data set;

[0007] S30: Extract features from the denoised and filled data set to obtain spectral features, time-domain features, and energy features;

[0008] S40: Perform feature fusion on the spectral features, time-domain features, and energy features to obtain a fused feature vector, and select some of the fused feature vectors to be divided into a training set and a test set;

[0009] S50: Construct a neural network classification model, train the neural network classification model with the training set to obtain a state classification model, optimize the state classification model with the test set, and input the fusion feature vector into the state classification model for classification;

[0010] S60: Select the data classified as abnormal state for detection and identify the abnormal state type;

[0011] S70: Select the data classified as normal state for health assessment, and comprehensively evaluate the health status of the low-voltage distribution network by calculating health key indicators and a health assessment function;

[0012] The original data set includes: power signals, power data, and power supply data;

[0013] The power signals include: current and voltage;

[0014] The power data includes: rated active power of the transformer, actual active power of the transformer, and three-phase load;

[0015] The power supply data includes: power supply radius, total power supply, total power sold, total power supply time, and voltage over-limit time;

[0016] The partial fusion feature vectors are randomly selected 40% from the fusion feature vectors.

[0017] Further, the S20 includes:

[0018] S21: Divide the missing data in the original data set into recently missing data and ordinary missing data, fill in the recently missing data by timestamp alignment interpolation, and fill in the ordinary missing data by Lagrange interpolation to obtain a filled data set;

[0019] S22: Denoise the filled data set, use a moving average filter to denoise the time-domain encoded signal, and use a Butterworth low-pass filter to denoise the frequency-domain encoded signal to obtain a denoised and filled data set.

[0020] Further, the S30 includes:

[0021] S31: Convert the power signal data in the denoised and filled data set from the time domain to the frequency domain representation through fast Fourier transform to obtain the energy distribution information at different frequencies;

[0022] S32: Perform spectral analysis on the energy distribution information and extract spectral features through the frequencies corresponding to the energy;

[0023] S33: Decompose the time series signals in the denoised and supplemented dataset into wavelet coefficients of different scales and frequencies through wavelet transform, calculate the mean, variance, skewness, and kurtosis of the wavelet coefficients to obtain time domain features;

[0024] S34: Calculate the total energy of the time domain signals in the denoised and supplemented dataset, and perform spectral density integration on the frequency domain signals in the denoised and supplemented dataset to obtain energy features.

[0025] Further, the S40 includes:

[0026] S41: Perform feature fusion on the spectral features, time domain features, and energy features to obtain a fused feature vector;

[0027] S42: Randomly divide the partial fused feature vector into a training set and a test set according to 8:2.

[0028] Further, the S50 includes:

[0029] S51: Determine the number of input layer nodes of the neural network classification model according to the dimension of the fused feature vector;

[0030] S52: Pass through three hidden layers, the hidden layers use fully connected layers, and the ReLU activation function is used;

[0031] S53: The output layer uses the Sigmoid activation function to divide the output results into abnormal states and normal states;

[0032] S54: Use the training set to train the neural network classification model to obtain a state classification model;

[0033] S55: Optimize the state classification model through the test machine;

[0034] S56: Input the fused feature vector into the state classification model to obtain abnormal state data and normal state data.

[0035] Further, the S60 includes:

[0036] S61: Define three types of abnormal state types according to the characteristics of the low-voltage distribution network, including: current mutation, frequency deviation, and voltage fluctuation;

[0037] S62: Calculate the local density of the abnormal state data and its neighborhood data using the local outlier factor, and average the ratio of the local density of the neighborhood data to the local density of the abnormal state data to obtain the LOF value of the abnormal state data;

[0038] S63: Identify the type that causes the abnormal state according to the LOF value of the abnormal state data.

[0039] Furthermore, the S70 includes:

[0040] S71: Define five key indicators according to the characteristics of the low-voltage distribution network, including: power supply radius, comprehensive line loss rate, transformer load rate, user voltage qualification rate, and three-phase load imbalance degree;

[0041] S72: Set the constraint conditions for the power supply radius of the low-voltage distribution line;

[0042] S73: Solve the comprehensive line loss rate of the low-voltage distribution network;

[0043] S74: Solve the transformer load rate of the low-voltage distribution network;

[0044] S75: Solve the voltage qualification rate of the low-voltage distribution network;

[0045] S76: Solve the three-phase load imbalance degree of the low-voltage distribution network;

[0046] S77: Calculate the weights of the five key indicators by the Lagrangian optimal multiplier method to obtain a health assessment function, and conduct a comprehensive health assessment of the low-voltage distribution network based on the normal state data according to the health assessment function.

[0047] The beneficial effects of the present invention are:

[0048] 1. The present invention provides a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal. By interpolating and denoising the collected original data set, noise and outliers are effectively removed, ensuring the accuracy and reliability of subsequent analysis, thereby improving the accuracy of monitoring and management of the low-voltage distribution network.

[0049] 2. The present invention provides a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal. By using the local outlier factor method for anomaly detection, abnormal states such as current mutations and frequency offsets can be accurately identified, which can help to timely discover and solve potential problems and ensure the stable operation of the low-voltage distribution network.

[0050] 3. The present invention provides a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal. By using the fast Fourier transform and wavelet transform to extract spectral, time-domain, and energy features and perform feature fusion, the accuracy and efficiency of state classification are improved, thus providing strong support for the intelligent management of the low-voltage distribution network.

[0051] 4. The present invention provides a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal. By calculating key indicators such as the comprehensive line loss rate, transformer load, and three-phase load imbalance degree, the health state of the distribution network is comprehensively evaluated, enabling management personnel to better grasp the operating state of the distribution network and improve the scientificity and efficiency of management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 FIG. is a schematic flow chart of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal according to the present invention.

[0053] Figure 2 FIG. is a schematic diagram of data preprocessing of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal according to the present invention.

[0054] Figure 3 FIG. is a schematic diagram of feature extraction of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal according to the present invention.

[0055] Figure 4 FIG. is a schematic diagram of feature fusion of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal according to the present invention.

[0056] Figure 5 FIG. is a schematic diagram of a state classification model of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal according to the present invention.

[0057] Figure 6 FIG. is a schematic diagram of abnormal state detection of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal according to the present invention.

[0058] Figure 7 FIG. is a schematic diagram of health assessment of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] The following describes in detail the preferred embodiments of the present invention with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a more clear and definite definition of the protection scope of the present invention.

[0060] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways different from those described herein; obviously, the embodiments in the specification are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0061] Please refer to Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 andFigure 7 , embodiments of the present invention include:

[0062] Please refer to Figure 1 , which is a schematic flow diagram of a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal provided by an embodiment of the present invention, and specifically may include: steps S10 to S70.

[0063] S10: The fusion terminal acquires the original data set of the low-voltage distribution network.

[0064] Further, the original data set includes: power signals, power data, power supply data;

[0065] The power signals include: current, voltage;

[0066] The power data includes: rated active power of the transformer, actual active power of the transformer, three-phase load;

[0067] The power supply data includes: power supply radius, total power supply, total power sold, total power supply time, voltage over-limit time;

[0068] The partial fusion feature vectors are randomly selected 40% from the fusion feature vectors.

[0069] Exemplarily, the original data set of the low-voltage distribution network can be acquired from the fusion terminal. The original data set includes: current, voltage, rated active power of the transformer, actual active power of the transformer, three-phase load, power supply radius, total power supply, total power sold, total power supply time, voltage over-limit time, and form a data set.

[0070] S20: Perform data preprocessing on the original data set to obtain a denoised and filled data set.

[0071] Please refer to Figure 2 , which is a schematic diagram of data preprocessing for a method for detecting the operating state of a low-voltage distribution network based on a fusion terminal provided by an embodiment of the present invention, and specifically may include: steps S21 to S22.

[0072] S21: Divide the missing data in the original data set into recent missing data and ordinary missing data. Perform timestamp alignment interpolation to fill the recent missing data, and perform Lagrange interpolation to fill the ordinary missing data to obtain a filled data set;

[0073] S22: Perform data denoising on the filled data set. Use a moving average filter to denoise the time-domain encoded signal, and use a Butterworth low-pass filter to denoise the frequency-domain encoded signal to obtain a denoised and filled data set;

[0074] Exemplarily, the missing data in the original dataset is divided into recent missing data and ordinary missing data, and the recent missing data and the ordinary missing data are filled according to various existing data filling methods to obtain a filled dataset, such as: K-Nearest Neighbors, matrix decomposition, interpolation method, RandomForest, Generative Adversarial Networks (GANs), etc.

[0075] Exemplarily, by using the interpolation method to fill the recent missing data and the ordinary missing data, the filled dataset can be obtained. The specific steps of the interpolation method are as follows:

[0076] S211: Perform timestamp alignment interpolation filling on the recent missing data;

[0077] The calculation formula for the timestamp alignment interpolation is:

[0078] ;

[0079] where is the inserted value at time , is the known value at timestamp , is the known value at timestamp ;

[0080] S212: Perform Lagrange interpolation filling on the ordinary missing data;

[0081] The calculation formula for the Lagrange interpolation filling is:

[0082] ;

[0083] ;

[0084] where is the interpolation result, is the Lagrange basis function, is the degree of the interpolation polynomial, is the th function value of the data point, is the th value of the data point, is the th value of the data point;

[0085] Exemplarily, by using the filtering and denoising method to denoise the filled dataset, the denoised filled dataset can be obtained. The specific steps of the filtering and denoising method are as follows:

[0086] S221: Denoise the time-domain encoded signal using a moving average filter;

[0087] The calculation formula of the moving average filter is:

[0088] y [ i ]= 1 MP ∑ j =0 MP -1 x i + j ;

[0089] Wherein, y [ i ] is the output signal, is the input signal, is the number of data points for averaging;

[0090] S222: Denoise the frequency-domain encoded signal using a Butterworth low-pass filter;

[0091] The calculation formula of the Butterworth low-pass filter is:

[0092] ;

[0093] Wherein, is the frequency response relationship between the input signal and the output signal, is the complex frequency-domain variable, is the cut-off frequency, is the order of the Butterworth low-pass filter;

[0094] S30: Extract features from the denoised and supplemented data set to obtain spectral features, time-domain features, and energy features.

[0095] Please refer to Figure 3 , which is a schematic diagram of feature extraction for a low-voltage distribution network operation status detection method based on a fusion terminal provided by an embodiment of the present invention, and specifically may include: steps S31 to S34.

[0096] S31: Convert the power signal data in the denoised and supplemented data set from the time domain to the frequency domain representation through fast Fourier transform to obtain the energy distribution information at different frequencies;

[0097] S32: Perform spectral analysis on the energy distribution information and extract spectral features by energy corresponding frequencies;

[0098] S33: Decompose the time series signals in the denoised and complemented dataset into wavelet coefficients of different scales and frequencies through wavelet transform, and calculate the mean, variance, skewness, and kurtosis of the wavelet coefficients to obtain time domain features;

[0099] S34: Calculate the total energy of the time domain signals in the denoised and complemented dataset, and perform spectral density integration on the frequency domain signals in the denoised and complemented dataset to obtain energy features;

[0100] Exemplarily, convert the power signal data in the denoised and complemented dataset into a frequency domain representation through fast Fourier transform. The fast Fourier transform calculation formula is: , where is the spectral value at frequency , is the th data point in the input sequence, is the sequence length, is the complex term. Obtain the energy distribution information at different frequencies, perform spectral analysis on the energy distribution information, and extract spectral features through the frequencies corresponding to the energy.

[0101] Exemplarily, decompose the time series signals in the denoised and complemented dataset into wavelet coefficients of different scales and frequencies through wavelet transform. The wavelet transform calculation formula is: , where is the wavelet coefficient, is the input signal, is the complex conjugate of the wavelet basis function, is the time variable, is the scale parameter, is the translation parameter. Calculate the mean, variance, skewness, and kurtosis of the wavelet coefficients to obtain time domain features.

[0102] Exemplarily, calculate the total energy of the time domain signals in the denoised and complemented dataset. The total energy calculation formula of the time domain signal is: , where is the total energy of the time domain signal, is the value of the th sampling point in the time domain signal, is the number of sampling points of the signal. Perform spectral density integration on the frequency domain signals in the denoised and complemented dataset. The spectral density integration calculation formula is: , where is the total energy of the frequency domain signal, is the spectral density at frequency , and are the frequency ranges of the integration.

[0103] S40: Perform feature fusion on the spectral features, time-domain features, and energy features to obtain a fused feature vector, and select some of the fused feature vectors to be divided into a training set and a test set.

[0104] Please refer to Figure 4 , which is a schematic diagram of feature fusion for a low-voltage distribution network operation status detection method provided by an embodiment of the present invention, and specifically may include: steps S41 to S42.

[0105] S41: Perform feature fusion on the spectral features, time-domain features, and energy features to obtain a fused feature vector;

[0106] S42: Randomly divide the partial fused feature vectors into a training set and a test set according to 8:2;

[0107] Exemplarily, perform feature fusion on the extracted spectral features, time-domain features, and energy features to obtain a fused feature vector, such as: Concatenation, Weighted Average, Cross, Stacking, Voting, Feature Selection, etc.

[0108] Exemplarily, perform feature fusion on the extracted spectral features, time-domain features, and energy features through concatenation to obtain a fused feature vector, and the fused feature vector is: X =[ X spectral , X time , X energy ] , where is the fused feature vector, is the spectral feature, is the time-domain feature, is the energy feature.

[0109] S50: Construct a neural network classification model, train the neural network classification model through the training set to obtain a status classification model, optimize the status classification model through the test set, and input the fused feature vector into the status classification model for classification.

[0110] Please refer to Figure 5 , which is a schematic diagram of the status classification model for a low-voltage distribution network operation status detection method provided by an embodiment of the present invention, and specifically may include: steps S51 to S56.

[0111] S51: Determine the number of input layer nodes of the neural network classification model according to the dimension of the fusion feature vector;

[0112] S52: Pass through three hidden layers, the hidden layers adopt fully connected layers, and the ReLU activation function is used;

[0113] S53: The output layer uses the Sigmoid activation function to divide the output results into abnormal states and normal states;

[0114] S54: Use the training set to train the neural network classification model to obtain a state classification model;

[0115] S55: Optimize the state classification model through the testing machine;

[0116] S56: Input the fusion feature vector into the state classification model to obtain abnormal state data and normal state data;

[0117] Exemplarily, a state classification model is obtained by establishing and training a neural network classification model. Determine the number of input layer nodes of the neural network classification model according to the dimension of the fusion feature vector. Pass through three hidden layers, the hidden layers adopt fully connected layers, and the ReLU activation function is used. The ReLU activation function is: , where is the input value, is to take the maximum value of 0 and , the output layer uses the Sigmoid activation function, and the Sigmoid activation function is , where is the input value. Through the Sigmoid activation function for binary classification, the output results are divided into abnormal states and normal states. Use the training set to train the neural network classification model to obtain a state classification model. Optimize the state classification model through the testing machine. Input the fusion feature vector into the state classification model to obtain abnormal state data and normal state data.

[0118] S60: Select the data classified as abnormal state data for detection and identify the abnormal state type.

[0119] Please refer to Figure 6 , which is a schematic diagram of abnormal state detection for a low-voltage distribution network operation state detection method based on a fusion terminal provided by an embodiment of the present invention, and specifically may include: steps S61 to S63.

[0120] S61: Define three abnormal state types according to the characteristics of the low-voltage distribution network, including: current mutation, frequency deviation, voltage fluctuation;

[0121] S62: Calculate the local density of the abnormal state data and its neighborhood data using the local outlier factor, and calculate the average value of the ratio of the local density of the neighborhood data to the local density of the abnormal state data to obtain the LOF value of the abnormal state data;

[0122] S63: Identify the type of abnormal state based on the LOF value of the abnormal state data;

[0123] Exemplarily, according to the characteristics of the low-voltage distribution network, the abnormal state types are divided into three cases, including: current mutation, frequency deviation, and voltage fluctuation. Calculate the local density of the abnormal state data and its neighborhood data using the local outlier factor. The local density calculation formula is: , where is the data point for which the local density is to be calculated, is the neighboring data point of the data point , is the data point to distance, is the local reachability density of the data point . Take the maximum value, calculate the average value of the ratio of the local density of the neighborhood data to the local density of the abnormal state data to obtain the LOF value of the abnormal state data. The LOF value calculation formula is: , where is the set of neighborhood data points of the data point , is the local reachability density of the data point . Identify the type of abnormal state based on the LOF value of the abnormal state data.

[0124] S70: Select the data classified as normal state data for health assessment, and comprehensively evaluate the health state of the low-voltage distribution network by calculating the health key indicators and the health assessment function.

[0125] Please refer to Figure 7 , which is the health assessment schematic diagram of a low-voltage distribution network operation state detection method provided by an embodiment of the present invention, and specifically may include: steps S71 to S77.

[0126] S71: Define five key indicators according to the characteristics of the low-voltage distribution network, including: power supply radius, comprehensive line loss rate, transformer load rate, user voltage qualification rate, and three-phase load imbalance degree;

[0127] S72: Set the power supply radius constraint conditions of the low-voltage distribution line;

[0128] S73: Solve the comprehensive line loss rate of the low-voltage distribution network;

[0129] S74: Solve the transformer load rate of the low-voltage distribution network;

[0130] S75: Solve the voltage qualification rate of the low-voltage distribution network;

[0131] S76: Solve the three-phase load imbalance of the low-voltage distribution network;

[0132] S77: Calculate the weights of the five key indicators by the Lagrangian optimal multiplier method to obtain a health assessment function, and conduct a comprehensive health assessment of the low-voltage distribution network based on the normal state data according to the health assessment function;

[0133] Exemplarily, five key indicators are defined according to the characteristics of the low-voltage distribution network, including: power supply radius, comprehensive line loss rate, transformer load, user voltage qualification rate, and three-phase load imbalance. Set the constraint conditions for the power supply radius of the low-voltage distribution line. The constraint conditions for the power supply radius are: , where is the power supply radius. Solve the comprehensive line loss rate of the low-voltage distribution network. The calculation formula for the comprehensive line loss rate is: , where is the comprehensive line loss rate, is the total power supply of the target substation area, is the total power sold. Solve the transformer load rate of the low-voltage distribution network. The calculation formula for the transformer load rate is: , where is the transformer load rate, is the rated active power of the transformer, is the actual active power of the transformer. Solve the voltage qualification rate of the low-voltage distribution network. The calculation formula for the voltage qualification rate is: , where is the voltage qualification rate, is the total power supply time, is the voltage over-limit time. Solve the three-phase load imbalance of the low-voltage distribution network. The calculation formula for the three-phase load imbalance is: , where is the three-phase load imbalance, is to select the maximum value, is the phase load at the low-voltage side outlet of the substation area, is the phase load at the low-voltage side outlet of the substation area, is the phase load at the low-voltage side outlet of the substation area. Calculate the weights of the five key indicators by the Lagrangian optimal multiplier method to obtain a health assessment function, and conduct a comprehensive health assessment of the low-voltage distribution network based on the normal state data according to the health assessment function.

[0134] The above are only embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. A method for detecting the operating status of a low-voltage distribution network based on a fusion terminal, characterized in that: include: S10: The fusion terminal obtains the original data set of the low-voltage distribution network; S20: performing data preprocessing on the original data set to obtain a denoised and padded data set, including: S21: dividing the missing data in the original data set into recent missing data and common missing data, using timestamp alignment interpolation to fill the recent missing data, and using Lagrange interpolation to fill the common missing data, to obtain a filled data set; S22: performing data denoising on the padded data set, denoising the time domain coded signal by using a moving average filter, and denoising the frequency domain coded signal by using a Butterworth low-pass filter, to obtain a denoised padded data set; The moving average filter calculation formula is: ； in, is the output signal, is the input signal, is the number of data points to be averaged; The Butterworth low-pass filter calculation formula is: ； in, is the frequency response relationship between the input signal and the output signal, is a complex frequency domain variable, is the cut-off frequency, is the order of the Butterworth low-pass filter; S30: extracting features from the denoising and padding data set to obtain spectrum features, time domain features, and energy features; S40: performing feature fusion on the spectrum features, time domain features, and energy features by cascading, weighted averaging, crossover, stacking, voting, or feature selection to obtain a fused feature vector, wherein the fused feature vector includes the fused spectrum features, time domain features, and energy features, and selecting part of the fused feature vector to be divided into a training set and a test set; S50: constructing a neural network classification model, training the neural network classification model through the training set to obtain a state classification model, optimizing the state classification model through the test set, and inputting the fused feature vector into the state classification model for classification; S60: Selecting data classified as abnormal state for detection and identifying the type of abnormal state, including: S61: According to the characteristics of low-voltage distribution network, three types of abnormal conditions are defined, including: current mutation, frequency deviation, and voltage fluctuation; S62: Calculate the local density of the abnormal state data and its neighborhood data using a local abnormal factor, average the ratio of the local density of the neighborhood data to the local density of the abnormal state data, and obtain the LOF value of the abnormal state data; The local density calculation formula is: ,in, is the data point for which the local density is to be calculated, For data points The neighboring data points of For data points arrive The distance For data points The local reachable density of The LOF value calculation formula is: ,in, For data points The set of neighborhood data points, For data points The local reachable density of S63: Identify the type causing the abnormal state according to the LOF value of the abnormal state data; S70: Select data classified as normal status for health assessment, and comprehensively evaluate the health status of the low-voltage distribution network by calculating key health indicators and health assessment functions; The original data set includes: power signal, power data, and power supply data; The power signal includes: current and voltage; The power data includes: rated active power of transformer, actual active power of transformer, and three-phase load; The power supply data includes: power supply radius, total power supply, total power sales, total power supply time, and voltage limit exceeding time; The partial fusion feature vectors are 40% randomly selected from the fusion feature vectors; The classification includes: abnormal state data, normal state data; The abnormal state types include: current mutation, frequency deviation, and voltage fluctuation.

2. A method for detecting the operating status of a low-voltage distribution network based on a fusion terminal according to claim 1, characterized in that: The S30 includes: S31: converting the power signal data in the denoised and padded data set from the time domain to the frequency domain through fast Fourier transform to obtain energy distribution information at different frequencies; S32: Performing spectrum analysis on the energy distribution information, and extracting spectrum features according to energy corresponding frequencies; S33: decomposing the time series signal in the denoising and padding data set into wavelet coefficients of different scales and frequencies through wavelet transform, calculating the mean, variance, skewness, and kurtosis of the wavelet coefficients, and obtaining time domain features; S34: Calculate the total signal energy of the time domain signal in the denoised and padded data set, and perform spectrum density integration on the frequency domain signal in the denoised and padded data set to obtain energy features.

3. A method for detecting the operating status of a low-voltage distribution network based on a fusion terminal according to claim 1, characterized in that: The S40 comprises: S41: performing feature fusion on the spectrum features, time domain features, and energy features to obtain a fused feature vector; S42: Randomly divide the partial fusion feature vectors into a training set and a test set according to a ratio of 8:

2.

4. A method for detecting the operating status of a low-voltage distribution network based on a fusion terminal as claimed in claim 1, characterized in that: The S50 comprises: S51: Determine the number of input layer nodes of the neural network classification model according to the dimension of the fused feature vector; S52: passing through three hidden layers, wherein the hidden layers are fully connected layers and use a ReLU activation function; S53: The output layer uses the Sigmoid activation function to divide the output results into abnormal state and normal state; S54: using the training set to train the neural network classification model to obtain a state classification model; S55: Optimizing the state classification model through the test set; S56: Input the fused feature vector into a state classification model to obtain abnormal state data and normal state data.

5. A method for detecting the operating status of a low-voltage distribution network based on a fusion terminal according to claim 1, characterized in that: The S70 includes: S71: Five key indicators are defined based on the characteristics of low-voltage distribution networks, including: power supply radius, comprehensive line loss rate, transformer load rate, user voltage qualification rate, and three-phase load imbalance; S72: Setting power supply radius constraint conditions of low voltage distribution lines; S73: solving the comprehensive line loss rate of the low voltage distribution network; S74: solving the transformer load factor of the low voltage distribution network; S75: Calculate the voltage qualification rate of the low voltage distribution network; S76: solving the three-phase load imbalance of the low-voltage distribution network; S77: Calculate the weights of the five key indicators using the Lagrange optimal multiplier method to obtain a health assessment function, and perform a comprehensive health assessment of the low-voltage distribution network on the normal state data based on the health assessment function.

Images