Method and device for predicting ferromagnetic resonance fault of power distribution network

Abstract

The invention provides a method and a device for predicting a ferromagnetic resonance fault of a power distribution network. The method comprises the following steps: acquiring electrical quantity information before and after the ferromagnetic resonance fault, and preprocessing the electrical quantity information; generating a fault information sequence according to the preprocessed electrical quantity information; the maximum Lyapunov index, the correlation dimension and the Kolmogorov entropy are calculated by means of the fault information sequence; and generating a ferromagnetic resonance fault prediction model according to the maximum Lyapunov index, the correlation dimension and the Kolmogorov entropy. According to the power distribution network ferromagnetic resonance fault prediction method and device provided by the invention, the predictability of determining the fault signal is realized, the identification speed and prediction accuracy of the ferromagnetic resonance fault are remarkably improved, technical support is provided for safe and stable operation of a power system, and the reliability and safety of the power system can be remarkably improved.

Description

Technical Field

[0001] The present application relates to the technical field of power system fault detection, and in particular to a method and device for predicting ferromagnetic resonance faults in a distribution network. Background Art

[0002] With the expansion of the scale and complexity of the distribution network, a single-phase grounding fault in a small current grounding system is prone to cause ferromagnetic resonance overvoltage, posing a serious threat to the insulation of equipment and systems. At present, scholars have conducted a lot of research on ferromagnetic resonance fault diagnosis methods, such as wavelet analysis, signal injection method, mathematical morphology, rough set method, chaos theory and atomic decomposition method. However, these methods still have problems of inaccurate recognition or slow response speed in practical applications.

[0003] In particular, the application of chaos theory in ferromagnetic resonance fault identification, although there is a prediction method based on the maximum Lyapunov exponent, this method mainly relies on the chaotic characteristics of a single dimension, which may limit its prediction accuracy and scope of application. Summary of the invention

[0004] In view of the problems in the prior art, the embodiments of the present application provide a method and device for predicting ferromagnetic resonance faults in a distribution network, which can at least partially solve the problems in the prior art.

[0005] In a first aspect, the present application provides a method for predicting ferromagnetic resonance faults in a distribution network, comprising:

[0006] Acquiring electrical quantity information before and after a ferromagnetic resonance fault occurs and preprocessing the electrical quantity information;

[0007] Generate a fault information sequence according to the preprocessed electrical quantity information;

[0008] Calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence;

[0009] A ferromagnetic resonance fault prediction model is generated according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy.

[0010] Further, generating a fault information sequence according to the preprocessed electrical quantity information includes:

[0011] Constructing time series based on preprocessed electrical quantity information;

[0012] The time series is reconstructed in phase space to obtain a fault information sequence.

[0013] Furthermore, generating a ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy includes:

[0014] The maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy are integrated to obtain a comprehensive chaotic feature;

[0015] Model training is performed on the comprehensive chaotic characteristics to generate a ferromagnetic resonance fault prediction model.

[0016] Furthermore, it also includes:

[0017] Obtain real-time electrical quantity information of the distribution network;

[0018] generating a real-time fault information sequence according to the real-time electrical quantity information;

[0019] The ferromagnetic resonance fault prediction model and the real-time fault information sequence are used to predict the ferromagnetic resonance fault of the distribution network.

[0020] In a second aspect, the present application provides a distribution network ferromagnetic resonance fault prediction device, comprising:

[0021] A fault information sequence generating unit, used for generating a fault information sequence according to the preprocessed electrical quantity information;

[0022] A chaos characteristic calculation unit, used for calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence;

[0023] A fault prediction model generating unit, used for generating a ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy;

[0024] The fault prediction unit is used to predict the ferromagnetic resonance fault of the distribution network by using the ferromagnetic resonance fault prediction model.

[0025] Furthermore, the fault information sequence generating unit includes:

[0026] A time series construction module is used to construct a time series based on the preprocessed electrical quantity information;

[0027] The fault information sequence generation module is used to perform phase space reconstruction on the time series to obtain a fault information sequence.

[0028] Furthermore, the fault prediction model generating unit includes:

[0029] A comprehensive chaos feature generation module, used for fusing the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy to obtain a comprehensive chaos feature;

[0030] The fault prediction model generation module is used to perform model training on the comprehensive chaotic characteristics to generate a ferromagnetic resonance fault prediction model.

[0031] Furthermore, it also includes:

[0032] An information acquisition unit, used to acquire electrical quantity information before and after a ferromagnetic resonance fault occurs;

[0033] The information preprocessing unit is used to preprocess the electrical quantity information before and after the ferromagnetic resonance fault occurs.

[0034] In a third aspect, the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any one of the above embodiments when executing the computer program.

[0035] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented.

[0036] In a fifth aspect, the present application provides a computer program product, wherein the computer program product includes a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented.

[0037] The present application provides a method and device for predicting ferromagnetic resonance faults in a distribution network, which obtains electrical quantity information before and after a ferromagnetic resonance fault occurs and preprocesses the electrical quantity information; generates a fault information sequence based on the preprocessed electrical quantity information; uses the fault information sequence to calculate the maximum Lyapunov exponent, correlation dimension, and Kolmogorov entropy; generates a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension, and the Kolmogorov entropy, thereby achieving the predictability of determining a fault signal, significantly improving the identification speed and prediction accuracy of ferromagnetic resonance faults, providing technical support for the safe and stable operation of a power system, and being able to significantly improve the reliability and safety of the power system.

[0038] Among them, by constructing and analyzing the time series of ferromagnetic resonance fault signals, the chaos judgment of the fault signal is realized; by reconstructing the time series in phase space, the fault information sequence is obtained, and key parameters such as the saturated correlation dimension are obtained; by using the fault information sequence to calculate the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy and generating a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy, the prediction accuracy and robustness can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 It is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application;

[0041] Figure 2 It is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application;

[0042] Figure 3 It is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application;

[0043] Figure 4 It is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application;

[0044] Figure 5 It is a structural schematic diagram of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application;

[0045] Figure 6 It is a structural schematic diagram of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application;

[0046] Figure 7 It is a structural schematic diagram of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application;

[0047] Figure 8 It is a structural schematic diagram of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application;

[0048] Fig. 9 It is a schematic diagram of the physical structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. Here, the illustrative embodiments of the present application and their descriptions are used to explain the present application, but are not intended to limit the present application. It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined arbitrarily with each other.

[0050] The following uses a server as an example of an execution subject to illustrate the specific implementation process of the distribution network ferromagnetic resonance fault prediction method provided in an embodiment of the present application.

[0051] Figure 1 FIG. 1 is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application. Figure 1 As shown, the distribution network ferromagnetic resonance fault prediction method provided by the present application includes:

[0052] S101: Acquire electrical quantity information before and after a ferromagnetic resonance fault occurs and pre-process the electrical quantity information;

[0053] S102: generating a fault information sequence according to the preprocessed electrical quantity information;

[0054] S103: Calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence;

[0055] S104: Generate a ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy.

[0056] from Figure 1 It can be seen from the shown process that the present application provides a method for predicting ferromagnetic resonance faults in distribution networks, by obtaining electrical quantity information before and after the occurrence of a ferromagnetic resonance fault and preprocessing the electrical quantity information; generating a fault information sequence based on the preprocessed electrical quantity information; using the fault information sequence to calculate the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy; generating a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy, thereby achieving the predictability of determining the fault signal, significantly improving the identification speed and prediction accuracy of ferromagnetic resonance faults, providing technical support for the safe and stable operation of the power system, and being able to significantly improve the reliability and safety of the power system.

[0057] Each step is explained in detail below.

[0058] S101: Acquire electrical quantity information before and after a ferromagnetic resonance fault occurs and pre-process the electrical quantity information;

[0059] Specifically, the server first obtains the electrical quantity information after the ferromagnetic resonance fault occurs under different working conditions and environments in the historical data, and the electrical quantity information includes zero-sequence voltage and zero-sequence current, etc. The server's preprocessing operations on the electrical quantity information before and after the ferromagnetic resonance fault include data cleaning, noise filtering, data alignment and synchronization, feature separation and normalization, etc.

[0060] In one embodiment, the data cleaning operation includes supplementing the missing data that may exist in the collected electrical quantity information by linear interpolation, Lagrange interpolation or filling method; detecting and eliminating abnormal data by statistical methods or machine learning models (such as isolation forest algorithm). The noise filtering operation includes applying digital filtering technology to the electrical quantity signal to remove high-frequency noise and power frequency interference. The data alignment and synchronization operation includes aligning the timestamps of the electrical quantity data collected from multiple channels to ensure data synchronization; resampling the signals with different sampling rates to unify the time scale. The feature separation operation includes extracting the characteristic frequency components of ferromagnetic resonance using spectrum analysis; segmenting the signals before and after the fault and marking the change interval of the electrical quantity signal. The normalization processing operation includes normalizing the data of different physical quantities, eliminating the calculation errors caused by different dimensions, and normalizing the data.

[0061] In one embodiment, smart sensors are deployed at key nodes of the distribution network to collect electrical quantity information in real time and set fault criteria, such as voltage surge, current mutation or harmonic content surge. Once the criteria are met, the data acquisition module is triggered to save the fault-related electrical quantity information.

[0062] S102: generating a fault information sequence according to the preprocessed electrical quantity information;

[0063] Specifically, the server generates a time series reflecting the dynamic behavior of the system based on the preprocessed electrical quantity information, where the electrical quantity information includes zero-sequence voltage and zero-sequence current after the occurrence of ferromagnetic resonance fault under different working conditions and environments. By preprocessing the electrical quantity information, the server can remove noise from the signal, ensure the accuracy and reliability of the signal, and facilitate subsequent analysis. The server generates a fault information sequence based on the preprocessed electrical quantity information, separates the main frequency components of the electrical signal based on the electrical quantity information, and extracts specific fault indicators to form time series data as the basis of the fault information sequence.

[0064] Figure 2 FIG. 1 is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application. Figure 2 As shown, S102 includes:

[0065] S201: constructing a time series according to the preprocessed electrical quantity information;

[0066] Specifically, the server selects fault-sensitive electrical quantity parameters from the preprocessed data based on the characteristics of ferromagnetic resonance. The sampled electrical quantity information is arranged into a time series in chronological order. A fixed time window or sliding window is set to segment the data to capture dynamic changes. For signal data with uneven sampling, interpolation methods are used to fill in gaps to ensure the continuity of the time series. Signals with different sampling rates are resampled to unify the time step to make the time series consistent. Detrending technology is used to remove long-term trend items in the signal, highlight the dynamic changes before and after the fault, and reduce noise interference through smoothing algorithms.

[0067] The server formats the constructed time series into a standard data structure and stores it in a database or memory to provide fast access for subsequent phase space reconstruction.

[0068] S202: Reconstruct the time series in phase space to obtain a fault information sequence.

[0069] Specifically, after constructing a time series according to the preprocessed electrical quantity information, the server reconstructs the time series in phase space to obtain a fault information sequence.

[0070] In one embodiment, the server determines the delay time and the embedding dimension, constructs a phase space trajectory matrix according to the delay time and the embedding dimension, extracts the fault information sequence from the phase space trajectory matrix, draws the phase space trajectory in two-dimensional or three-dimensional space, observes whether the trajectory exhibits chaotic attractor characteristics, and determines the dynamic behavior of the system.

[0071] The delay time can be determined by calculating the autocorrelation coefficient of the time series and selecting the time point when the autocorrelation coefficient approaches zero for the first time as the delay time, or by calculating the average mutual information of the time series and selecting the first minimum point of the mutual information value as the delay time, or by using a graphical method. The false nearest neighbor method (FNN) can be used to determine the embedding dimension. In the process of gradually increasing the embedding dimension, the proportion of false neighbor points is monitored to determine the reasonable minimum embedding dimension m.

[0072] According to the delay time τ and the embedding dimension m, the trajectory matrix is ​​constructed:

[0073] X={(x(t),x(t+τ),x(t+2τ),…,x(t+(m-1)τ))} (1)

[0074] Among them, x(t) represents the value of the time series at time t, and each row of the trajectory matrix corresponds to a point in the phase space.

[0075] By analyzing the trajectory matrix, calculating the Euclidean distance of adjacent points, identifying the significant characteristic changes of the signal, and extracting the sequence related to the ferromagnetic resonance fault, the trajectory matrix can be reduced in dimension by combining principal component analysis (PCA) to extract the main fault feature sequence.

[0076] In one embodiment, the time series data to be processed is input, and the server calculates the correlation integral C(m,N,r,τ) based on the determined embedding dimension m, the number of data points N, the neighborhood radius r, and the delay time τ to evaluate the correlation of the time series data. Using the correlation integral C(m,N,r,τ), the statistic S(m,N,r,τ) is constructed. The specific expression is S(m,N,r,τ)=-C(m,N,r,τ)+(m-1)C(m-1,N,r,τ), the purpose is to extract useful statistical information from the correlation integral. According to the statistic S(m,N,r,τ), △S(m,τ) is calculated, and the specific expression is △

[0077] S(m,τ)=S(m+1,τ)-S(m,τ) is used to evaluate the change of statistics under different embedding dimensions. Find the zero point on the curve of △S(m,τ). The embedding dimension m corresponding to this zero point is the optimal embedding dimension, and the delay time τ corresponding to this zero point is also the optimal delay time. According to the optimal delay time τ and embedding dimension m, construct the trajectory matrix, and extract the main fault feature sequence by analyzing the trajectory matrix.

[0078] S103: Calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence;

[0079] Specifically, the server uses the collected fault information sequence to accurately calculate and analyze multiple key chaotic characteristic parameters. Among them, the maximum Lyapunov exponent is an important indicator to measure the degree of chaos in the system and its sensitivity to initial conditions; the correlation dimension carefully describes the complexity of the distribution of fault electrical signals in the phase space from a geometric perspective; and the Kolmogorov entropy can reflect the degree of disorder and information generation rate of the system. The purpose of in-depth analysis of chaotic characteristic parameters is to comprehensively and deeply reflect the complex chaotic characteristics hidden in the fault electrical signals, provide reliable data support and theoretical basis for subsequent fault prediction, and help the stable and efficient operation of the entire electrical system.

[0080] In one embodiment, the method for calculating the maximum Lyapunov exponent, that is, the exponential divergence rate between adjacent trajectories is as follows: constructing a phase space trajectory matrix X according to the time series x(t), the delay time τ and the embedding dimension m; finding the closest trajectory point X in the trajectory matrix i and X j; Calculate the distance between adjacent points as the time step Δt increases; For all pairs of adjacent points, calculate the average divergence rate.

[0081] In one embodiment, the method for calculating the correlation dimension is: for all points in the trajectory matrix X i and X j , calculate the Euclidean distance r ij , count the proportion of points less than the threshold ∈; take lnC(∈) to perform a linear fit on ln∈, and the slope is the correlation dimension D2.

[0082] In one embodiment, the method for calculating Kolmogorov entropy is: divide the phase space where the trajectory point is located into boxes of equal length; count the frequency of the trajectory point entering each box to obtain the probability distribution P i , according to the box probability distribution, calculate the Shannon entropy; calculate the rate of change of the Shannon entropy of adjacent time steps Δt as the Kolmogorov entropy.

[0083] S104: generating a ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy;

[0084] Specifically, based on the above three chaos characteristic parameters, namely the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy, the server relies on the machine learning algorithm architecture, sets the chaos characteristic parameters as the input of the algorithm, and generates a ferromagnetic resonance fault prediction model through model training to predict whether the ferromagnetic resonance fault will occur and its severity. The server repeatedly analyzes, fits and optimizes these data in strict accordance with the rules and logic of the machine learning algorithm.

[0085] During multiple data iterations and model parameter adjustments, the server derived the correlation between ferromagnetic resonance faults and chaotic characteristic parameters, and finally generated a ferromagnetic resonance fault prediction model. The ferromagnetic resonance fault prediction model has a strong prediction capability. It can accurately predict whether a ferromagnetic resonance fault is likely to occur based on the input real-time chaotic characteristic parameters. At the same time, it can also make a scientific and reasonable assessment of the severity of the fault, providing a critical and reliable basis for power system operation and maintenance personnel to formulate response strategies in advance and reasonably allocate resources, thereby effectively ensuring the safe and stable operation of the power system and avoiding losses and risks caused by ferromagnetic resonance faults.

[0086] Figure 3 FIG. 1 is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application. Figure 3 As shown, S104 includes:

[0087] S301: fusing the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy to obtain a comprehensive chaotic feature;

[0088] Specifically, before fusing the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy, the server first normalizes the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy, performs data noise reduction, correlation analysis and independence analysis on the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy to eliminate the influence of dimensional differences, remove possible noise interference, and eliminate redundant or weakly correlated features.

[0089] The maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy can be fused using a linear fusion strategy or a nonlinear fusion strategy. The linear fusion strategy is to perform weighted averaging of each feature according to a fixed weight, and the nonlinear fusion strategy is to perform nonlinear fusion of features using a machine learning method.

[0090] In one embodiment, nonlinear fusion may use technical means such as PCA (Principal Component Analysis), kernel function or neural network.

[0091] S302: Performing model training on the comprehensive chaos characteristics to generate a ferromagnetic resonance fault prediction model.

[0092] Specifically, the server divides the data into training set, validation set and test set according to the comprehensive chaos feature sequence. The fault events are labeled, such as "normal state" and "ferromagnetic resonance fault state". The detailed classification can label different types of ferromagnetic resonance faults (such as single-phase fault, three-phase fault, etc.). The server inputs the comprehensive chaos features for model training, uses grid search or random search to optimize the model hyperparameters, and prevents overfitting.

[0093] In one embodiment, model training may use a support vector machine (SVM), a decision tree, or a deep learning model.

[0094] Figure 4 FIG. 1 is a flow chart of a method for predicting ferromagnetic resonance faults in a distribution network provided in an embodiment of the present application. Figure 4 As shown, it also includes:

[0095] S401: Acquire real-time electrical quantity information of the distribution network;

[0096] S402: generating a real-time fault information sequence according to the real-time electrical quantity information;

[0097] S403: Predicting distribution network ferromagnetic resonance faults using the ferromagnetic resonance fault prediction model and the real-time fault information sequence.

[0098] Specifically, the server accesses the electrical quantity information of each key node of the distribution network in real time through the data transmission link. The electrical quantity information includes zero-sequence voltage and zero-sequence current. After receiving the electrical quantity information, the server generates the latest fault information sequence reflecting the current operating status of the distribution network. Then, the server will extract the real-time chaotic characteristic parameters from the fault information sequence, and input the real-time chaotic characteristic parameters into the pre-built and fully trained ferromagnetic resonance fault prediction model, so that the ferromagnetic resonance fault prediction model can quickly perform calculations and analysis based on its internal complex mathematical model and logical rules, and then output the prediction results of ferromagnetic resonance faults.

[0099] Once the key indicators in the prediction results exceed the preset threshold, the server will automatically trigger the fault alarm mechanism and quickly generate a detailed fault alarm signal. The fault alarm signal not only contains the core information of the fault prediction, but also attaches the operating status data of the current distribution network and the key fragments of the fault information sequence, so that the operation and maintenance personnel can fully and deeply understand the fault situation. Subsequently, the server sends the alarm information to the operation and maintenance center to ensure that the relevant operation and maintenance personnel can obtain the fault alarm information as soon as possible. After receiving the fault prediction results, the operation and maintenance personnel will formulate and take a series of targeted preventive measures based on the fault prediction results.

[0100] In one embodiment, the prediction result includes occurrence probability, failure type and estimated time.

[0101] In one embodiment, preventive measures include adjusting loads, switching backup lines, and switching compensation equipment to reduce the risk of failures, eliminate potential failure hazards in advance, ensure that the impact on power supply to users is minimized when a failure occurs, and ensure the safe and stable operation of the distribution network.

[0102] The present application provides a method for predicting ferromagnetic resonance faults in a distribution network, which obtains electrical quantity information before and after a ferromagnetic resonance fault occurs and preprocesses the electrical quantity information; generates a fault information sequence based on the preprocessed electrical quantity information; uses the fault information sequence to calculate a maximum Lyapunov exponent, a correlation dimension, and a Kolmogorov entropy; generates a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension, and the Kolmogorov entropy, thereby achieving the predictability of determining a fault signal, significantly improving the identification speed and prediction accuracy of ferromagnetic resonance faults, providing technical support for the safe and stable operation of a power system, and being able to significantly improve the reliability and safety of the power system.

[0103] Among them, by constructing and analyzing the time series of ferromagnetic resonance fault signals, the chaos judgment of the fault signal is realized; by reconstructing the time series in phase space, the fault information sequence is obtained, and key parameters such as the saturated correlation dimension are obtained; by using the fault information sequence to calculate the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy and generating a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy, the prediction accuracy and robustness can be improved.

[0104] Based on the same inventive concept, the embodiments of the present application also provide a distribution network ferromagnetic resonance fault prediction device, which can be used to implement the method described in the above embodiments, as described in the following embodiments. Since the principle of solving the problem by the distribution network ferromagnetic resonance fault prediction device is similar to that of the distribution network ferromagnetic resonance fault prediction method, the implementation of the distribution network ferromagnetic resonance fault prediction device can refer to the implementation of the method based on software performance benchmark determination, and the repeated parts will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements predetermined functions. Although the system described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0105] Figure 5 is a schematic diagram of the structure of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application, such as Figure 5 As shown, the device comprises:

[0106] An information preprocessing unit 501 is used to obtain electrical quantity information before and after the ferromagnetic resonance fault occurs and preprocess the electrical quantity information;

[0107] Specifically, the information preprocessing unit 501 first obtains electrical quantity information after the occurrence of the ferromagnetic resonance fault under different working conditions and different environments in the historical data, and the electrical quantity information includes zero-sequence voltage and zero-sequence current, etc. The preprocessing operation of the information preprocessing unit 501 on the electrical quantity information before and after the ferromagnetic resonance fault occurs includes data cleaning, noise filtering, data alignment and synchronization, feature separation and normalization processing, etc.

[0108] In one embodiment, the data cleaning operation includes supplementing the missing data that may exist in the collected electrical quantity information by linear interpolation, Lagrange interpolation or filling method; detecting and eliminating abnormal data by statistical methods or machine learning models (such as isolation forest algorithm). The noise filtering operation includes applying digital filtering technology to the electrical quantity signal to remove high-frequency noise and power frequency interference. The data alignment and synchronization operation includes aligning the timestamps of the electrical quantity data collected from multiple channels to ensure data synchronization; resampling the signals with different sampling rates to unify the time scale. The feature separation operation includes extracting the characteristic frequency components of ferromagnetic resonance using spectrum analysis; segmenting the signals before and after the fault and marking the change interval of the electrical quantity signal. The normalization processing operation includes normalizing the data of different physical quantities, eliminating the calculation errors caused by different dimensions, and normalizing the data.

[0109] In one embodiment, smart sensors are deployed at key nodes of the distribution network to collect electrical quantity information in real time and set fault criteria, such as voltage surge, current mutation or harmonic content surge. Once the criteria are met, the data acquisition module is triggered to save the fault-related electrical quantity information.

[0110] A fault information sequence generating unit 502 is used to generate a fault information sequence according to the preprocessed electrical quantity information;

[0111] Specifically, the fault information sequence generation unit 502 generates a time series reflecting the dynamic behavior of the system based on the preprocessed electrical quantity information, wherein the electrical quantity information includes zero-sequence voltage and zero-sequence current after the occurrence of ferromagnetic resonance fault under different working conditions and environments. By preprocessing the electrical quantity information, the server can remove noise from the signal, ensure the accuracy and reliability of the signal, and facilitate subsequent analysis. The fault information sequence generation unit 502 generates a fault information sequence based on the preprocessed electrical quantity information, separates the main frequency components of the electrical signal based on the electrical quantity information, and extracts specific fault indicators to form time series data as the basis of the fault information sequence.

[0112] A chaos characteristic calculation unit 503, used for calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence;

[0113] Specifically, the chaos characteristic calculation unit 503 uses the collected fault information sequence to accurately calculate and analyze multiple key chaos characteristic parameters. Among them, the maximum Lyapunov exponent is an important indicator to measure the degree of chaos in the system and its sensitivity to initial conditions; the correlation dimension carefully describes the complexity of the distribution of the fault electrical signal in the phase space from a geometric perspective; and the Kolmogorov entropy can reflect the degree of disorder and information generation rate of the system. The purpose of in-depth analysis of the chaos characteristic parameters is to comprehensively and deeply reflect the complex chaotic characteristics hidden in the fault electrical signal, provide reliable data support and theoretical basis for subsequent fault prediction, and help the stable and efficient operation of the entire electrical system.

[0114] A fault prediction model generating unit 504 is used to generate a ferroresonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy;

[0115] Specifically, the fault prediction model generation unit 504 is based on the above three chaos characteristic parameters, namely, the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy, and relies on the machine learning algorithm architecture to set the chaos characteristic parameters as the input of the algorithm, and generates a ferromagnetic resonance fault prediction model through model training to predict whether the ferromagnetic resonance fault occurs and the severity. The fault prediction model generation unit 503 strictly follows the rules and logic of the machine learning algorithm to repeatedly analyze, fit and optimize these data.

[0116] In the process of multiple data iterations and model parameter adjustments, the fault prediction model generation unit 503 obtains the correlation between the ferromagnetic resonance fault and the electrical quantity information, and finally generates a ferromagnetic resonance fault prediction model. The ferromagnetic resonance fault prediction model has a strong prediction ability. It can accurately predict whether a ferromagnetic resonance fault may occur based on the input real-time electrical quantity information, and can also scientifically and reasonably evaluate the severity of the fault, which provides a very critical and reliable basis for the power system operation and maintenance personnel to formulate response strategies in advance and reasonably allocate resources, thereby effectively ensuring the safe and stable operation of the power system and avoiding losses and risks caused by ferromagnetic resonance faults.

[0117] Figure 6 is a schematic diagram of the structure of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application, Figure 5 Based on the embodiment, further, Figure 6 As shown, the distribution network ferromagnetic resonance fault prediction device provided by the present application also includes:

[0118] A time series construction module 601 is used to construct a time series according to the preprocessed electrical quantity information;

[0119] The fault information sequence generating module 602 is used to perform phase space reconstruction on the time sequence to obtain a fault information sequence.

[0120] Figure 7 is a schematic diagram of the structure of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application, Figure 5 Based on the embodiment, further, Figure 7 As shown, the distribution network ferromagnetic resonance fault prediction device provided by the present application also includes:

[0121] A comprehensive chaos feature generation module 701 is used to fuse the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy to obtain a comprehensive chaos feature;

[0122] The fault prediction model generation module 702 is used to perform model training on the comprehensive chaotic characteristics to generate a ferromagnetic resonance fault prediction model.

[0123] Figure 8 is a schematic diagram of the structure of a distribution network ferromagnetic resonance fault prediction device provided by an embodiment of the present application, Figure 5 Based on the embodiment, further, Figure 8 As shown, the distribution network ferromagnetic resonance fault prediction device provided by the present application also includes:

[0124] A real-time information acquisition unit 801 is used to acquire real-time electrical quantity information of the distribution network;

[0125] A real-time sequence generating unit 802, configured to generate a real-time fault information sequence according to the real-time electrical quantity information;

[0126] The fault prediction unit 803 is used to predict the ferromagnetic resonance fault of the distribution network by using the ferromagnetic resonance fault prediction model and the real-time fault information sequence.

[0127] The present application provides a method and device for predicting ferromagnetic resonance faults in a distribution network, which obtains electrical quantity information before and after a ferromagnetic resonance fault occurs and preprocesses the electrical quantity information; generates a fault information sequence based on the preprocessed electrical quantity information; uses the fault information sequence to calculate the maximum Lyapunov exponent, correlation dimension, and Kolmogorov entropy; generates a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension, and the Kolmogorov entropy, thereby achieving the predictability of determining a fault signal, significantly improving the identification speed and prediction accuracy of ferromagnetic resonance faults, providing technical support for the safe and stable operation of a power system, and being able to significantly improve the reliability and safety of the power system.

[0128] Among them, by constructing and analyzing the time series of ferromagnetic resonance fault signals, the chaos judgment of the fault signal is realized; by reconstructing the time series in phase space, the fault information sequence is obtained, and key parameters such as the saturated correlation dimension are obtained; by using the fault information sequence to calculate the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy and generating a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy, the prediction accuracy and robustness can be improved.

[0129] Fig. 9 is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application, such as Fig. 9 As shown, the electronic device may include: a processor (processor) 901, a communication interface (Communications Interface) 902, a memory (memory) 903 and a communication bus 904, wherein the processor 901, the communication interface 902, and the memory 903 complete mutual communication through the communication bus 904. The processor 901 may call the logic instructions in the memory 903 to execute the following method: obtaining electrical quantity information before and after the occurrence of a ferromagnetic resonance fault and preprocessing the electrical quantity information; generating a fault information sequence according to the preprocessed electrical quantity information; calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence; generating a ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy.

[0130] In addition, the logic instructions in the above-mentioned memory 903 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk.

[0131] The present embodiment discloses a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments, for example, including: obtaining electrical quantity information before and after a ferromagnetic resonance fault occurs and preprocessing the electrical quantity information; generating a fault information sequence according to the preprocessed electrical quantity information; calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence; generating a ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy.

[0132] The present embodiment provides a computer-readable storage medium, which stores a computer program. The computer program enables the computer to execute the methods provided by the above-mentioned method embodiments, for example, including: obtaining electrical quantity information before and after a ferromagnetic resonance fault occurs and preprocessing the electrical quantity information; generating a fault information sequence based on the preprocessed electrical quantity information; using the fault information sequence to calculate the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy; generating a ferromagnetic resonance fault prediction model based on the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy.

[0133] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0134] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0135] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0136] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0137] In the description of this specification, the description with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0138] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for predicting ferromagnetic resonance faults in a distribution network, characterized in that: include: Acquiring electrical quantity information before and after a ferromagnetic resonance fault occurs and preprocessing the electrical quantity information; Generate a fault information sequence according to the preprocessed electrical quantity information; Calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence; A ferromagnetic resonance fault prediction model is generated according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy.

2. The method for predicting ferromagnetic resonance faults in a power distribution network according to claim 1, characterized in that: The generating of the fault information sequence according to the preprocessed electrical quantity information comprises: Constructing time series based on preprocessed electrical quantity information; The time series is reconstructed in phase space to obtain a fault information sequence.

3. The method for predicting ferromagnetic resonance faults in a distribution network according to claim 1, characterized in that: The generating of the ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy comprises: The maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy are integrated to obtain a comprehensive chaotic feature; Model training is performed on the comprehensive chaotic characteristics to generate a ferromagnetic resonance fault prediction model.

4. The method for predicting ferromagnetic resonance faults in a power distribution network according to claim 1, characterized in that: Also includes: Obtain real-time electrical quantity information of the distribution network; generating a real-time fault information sequence according to the real-time electrical quantity information; The ferromagnetic resonance fault prediction model and the real-time fault information sequence are used to predict the ferromagnetic resonance fault of the distribution network.

5. A distribution network ferromagnetic resonance fault prediction device, characterized in that: include: An information preprocessing unit, used to obtain electrical quantity information before and after the ferromagnetic resonance fault occurs and preprocess the electrical quantity information; A fault information sequence generating unit, used for generating a fault information sequence according to the preprocessed electrical quantity information; A chaos characteristic calculation unit, used for calculating the maximum Lyapunov exponent, correlation dimension and Kolmogorov entropy using the fault information sequence; A fault prediction model generating unit is used to generate a ferromagnetic resonance fault prediction model according to the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy.

6. The distribution network ferromagnetic resonance fault prediction device according to claim 5, characterized in that: The fault information sequence generating unit comprises: A time series construction module is used to construct a time series based on the preprocessed electrical quantity information; The fault information sequence generation module is used to perform phase space reconstruction on the time series to obtain a fault information sequence.

7. The distribution network ferromagnetic resonance fault prediction device according to claim 5, characterized in that: The fault prediction model generating unit comprises: A comprehensive chaos feature generation module, used for fusing the maximum Lyapunov exponent, the correlation dimension and the Kolmogorov entropy to obtain a comprehensive chaos feature; The fault prediction model generation module is used to perform model training on the comprehensive chaotic characteristics to generate a ferromagnetic resonance fault prediction model.

8. The distribution network ferromagnetic resonance fault prediction device according to claim 5, characterized in that: Also includes: A real-time information acquisition unit, used to acquire real-time electrical quantity information of the distribution network; A real-time sequence generating unit, used for generating a real-time fault information sequence according to the real-time electrical quantity information; A fault prediction unit is used to predict the ferromagnetic resonance fault of the distribution network by using the ferromagnetic resonance fault prediction model and the real-time fault information sequence.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

11. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images