Power distribution network electric energy quality multi-disturbance source positioning and traceability analysis method, system and device and medium

Abstract

The invention relates to the technical field of power system power quality analysis and control, and discloses a power distribution network power quality multi-disturbance-source positioning and traceability analysis method, system and device and a medium, and the method comprises the steps: optimally configuring a monitoring point network according to a power distribution network topological structure; collecting three-phase voltage and current data of each monitoring point, and extracting multi-dimensional features after preprocessing; identifying disturbance types by adopting a hierarchical classification strategy, and separating composite disturbance; the disturbance information and the time sequence of each monitoring point are fused, and accurate space positioning of harmonic waves and transient disturbance sources is realized by combining parameters of the power distribution network; establishing a disturbance propagation model, and analyzing a propagation path and an influence range; and finally, generating a comprehensive traceability report containing a positioning result, disturbance characteristics, influence evaluation and governance suggestions, and calculating the overall traceability confidence. According to the method, the problems of accurate positioning and tracing in a multi-disturbance-source coexistence scene are effectively solved, and the positioning accuracy and the recognition reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of power quality analysis and control technology, and in particular to a method, system, equipment and medium for locating and tracing multiple disturbance sources in power quality of distribution networks. Background Technology

[0002] With the integration of numerous nonlinear loads, distributed power sources, and power electronic equipment into the distribution network, power quality disturbances are becoming increasingly prominent. Existing methods for locating and tracing the sources of power quality disturbances have the following main shortcomings: 1) The layout of monitoring points lacks optimization, resulting in insufficient accuracy in locating disturbance sources. Traditional power quality monitoring mainly relies on fixed monitoring points on the outgoing side of substations, which are limited in number and location. When multiple disturbance sources exist in the distribution network, the limited number of monitoring points cannot accurately capture all disturbance events, leading to significant errors in disturbance source location.

[0003] 2) The method for identifying disturbance types is limited, resulting in low accuracy when multiple types of disturbances coexist. Existing disturbance identification methods are mostly designed for single disturbance types, such as using Fourier transform to identify harmonics and using RMS sequence analysis to identify voltage sags. When multiple disturbances such as harmonics, voltage sags, voltage fluctuations, and interharmonics exist simultaneously in a distribution network, single feature extraction methods are difficult to accurately distinguish the disturbance types.

[0004] 3) Disturbance source localization relies on single-point information and lacks a multi-point data fusion mechanism. Existing location methods primarily rely on voltage or current information from a single monitoring point to pinpoint disturbance sources, such as estimating fault distance based on voltage sags or determining harmonic source locations based on harmonic current directions. This single-point location method is susceptible to measurement errors and inaccurate line parameters, resulting in unreliable location results. Summary of the Invention

[0005] In view of the above-mentioned existing problems, the present invention provides a method, system, equipment and medium for locating and tracing multiple disturbance sources in power quality of power distribution networks.

[0006] Therefore, the technical problem solved by the present invention is the technical defects of the existing technology, such as unreasonable layout of monitoring points, low accuracy of multi-type disturbance identification, poor reliability of single-point positioning, and lack of propagation analysis.

[0007] To address the aforementioned technical problems, this invention provides the following technical solution: a method for locating and tracing multiple sources of power quality disturbances in a distribution network, comprising: establishing an optimal configuration model for monitoring points based on the distribution network topology, line impedance parameters, and historical disturbance distribution through observability analysis to determine the optimal location and number of monitoring points; deploying power quality monitoring devices at the optimal monitoring points to collect three-phase voltage and current waveform data, and performing outlier detection and signal denoising processing to obtain a pre-processed clean signal; The preprocessed clean signal is subjected to time domain, frequency domain, and time-frequency domain feature extraction to construct a comprehensive feature vector. Based on the comprehensive feature vector, a hierarchical classification strategy is adopted to identify the disturbance type, and disturbance separation is performed in the scenario of multiple disturbances coexisting. The disturbance source location is located using a data fusion method. Based on the located disturbance source location, a disturbance propagation model is established to analyze the propagation path and impact range of the disturbance in the distribution network and generate a disturbance source tracing report.

[0008] As a preferred embodiment of the method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in this invention, the determination of the optimal monitoring point location and number includes: calculating the observability of each node based on the distribution network topology and line parameters using an observability analysis method; and using an optimization algorithm to minimize the number of monitoring devices, thereby obtaining an optimized monitoring point configuration scheme under the premise of meeting the observability threshold.

[0009] As a preferred embodiment of the method for locating and tracing multiple disturbance sources in power quality of distribution networks as described in this invention, the method for obtaining a pre-processed clean signal includes: collecting three-phase voltage and current waveform data by deploying power quality monitoring devices at each monitoring point, and processing the original data using anomaly detection algorithms and signal denoising techniques to obtain a clean signal for subsequent feature extraction.

[0010] As a preferred embodiment of the method for locating and tracing multiple disturbance sources in power quality of distribution networks according to the present invention, the construction of the comprehensive feature vector includes extraction of time-domain features, frequency-domain features, and time-frequency-domain features. The hierarchical classification strategy first distinguishes between steady-state disturbances and transient disturbances, then further identifies specific disturbance types, and employs a disturbance separation algorithm to decompose composite disturbance signals in scenarios with multiple disturbances coexisting.

[0011] As a preferred embodiment of the method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in this invention, the method of locating the disturbance source using data fusion includes performing time synchronization verification and ensuring that the same disturbance event is observed by comparing the disturbance detection time difference of each monitoring point. Define time synchronization criteria: In the formula, The disturbance detection time difference between monitoring points k and l. and The timestamps for the detection of disturbances (from step two). To determine the maximum allowable time difference, a sampling period is selected. Observational data from monitoring points that meet the time synchronization criteria are classified as the same event; For locating harmonic disturbance sources, the harmonic power direction method and harmonic impedance method are used. By calculating the harmonic active power and direction index at each monitoring point and combining it with the line harmonic impedance, the location of the harmonic source is determined. Calculate the active power of the h-th harmonic at monitoring point k: In the formula, The active power of the h-th harmonic is and These are the effective values ​​of harmonic voltage and current. For harmonic voltage and current phase; define harmonic direction index: In the formula, For harmonic direction, This indicates that harmonics flow from the node to the line. This indicates that harmonics flow from the line to the node; by analyzing the harmonic direction at each monitoring point, the range of harmonic source locations is narrowed down. For locating transient disturbance sources, the traveling wave time difference method is adopted. By calculating the time difference and propagation speed of the disturbance reaching different monitoring points, and combining the distribution network topology, the least squares method is used to solve for the coordinates of the disturbance source.

[0012] As a preferred embodiment of the method for locating and tracing multiple disturbance sources in power quality of distribution networks as described in this invention, the step of establishing a disturbance propagation model to analyze the propagation path and influence range includes, for harmonic disturbances, using the harmonic power flow calculation method, calculating the harmonic voltage distribution at each node of the distribution network through the harmonic impedance matrix and the harmonic source injection current, and defining the harmonic propagation attenuation coefficient to analyze the attenuation law. For transient disturbances, a traveling wave propagation model is adopted. The attenuation law of transient disturbances propagating on the line is as follows: In the formula, Let be the transient voltage amplitude at a distance d from the disturbance source. The transient voltage amplitude at the disturbance source. This refers to the attenuation coefficient; the attenuation coefficient is calculated based on the line parameters. In the formula, R is the resistance per unit length of the line. The characteristic impedance of the line. L and C are the inductance and capacitance per unit length of the line, respectively; Calculate the propagation amplitude of the disturbance along each line and identify the affected lines; define the criteria for determining the scope of the disturbance's impact. In the formula, For the set of affected nodes, and Let be the harmonic voltage and transient voltage at node i, respectively. To influence the threshold.

[0013] As a preferred embodiment of the method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in this invention, the generation of the disturbance source tracing report includes: calculating the comprehensive tracing confidence level, which is obtained by weighting the disturbance type identification confidence level, the location confidence level, and the propagation analysis confidence level. Generate a source tracing report, including the location coordinates of the disturbance source, the type of disturbance, characteristic parameters, the scope of impact, and mitigation recommendations; Based on the type and degree of disturbance, suggestions are given for calculating the filter capacity or dynamic voltage restorer capacity.

[0014] This invention provides a system for locating and tracing multiple sources of power quality disturbances in power distribution networks.

[0015] As a preferred embodiment of the power quality multi-disturbance source location and tracing analysis system for power distribution networks described in this invention, it includes: a monitoring point configuration module, a data acquisition and preprocessing module, a feature extraction and identification module, a location and analysis module, and a tracing output module; The monitoring point configuration module is used to establish an optimal monitoring point configuration model based on the distribution network topology, line impedance parameters, and historical disturbance distribution, and to determine the optimal monitoring point location and number. The data acquisition and preprocessing module acquires three-phase voltage and current waveform data, performs outlier detection and signal denoising, and obtains a preprocessed clean signal. The feature extraction and recognition module is used to extract time-domain, frequency-domain, and time-frequency-domain features from the preprocessed clean signal, construct a comprehensive feature vector, and identify the disturbance type based on the comprehensive feature vector using a hierarchical classification strategy. The positioning and analysis module is used to integrate the disturbance identification results and time information of each monitoring point, use data fusion methods to locate the disturbance source, and establish a disturbance propagation model to analyze the propagation path and impact range of the disturbance in the power distribution network. The source tracing output module is used to generate a disturbance source tracing report, including the location, type, impact range, and governance recommendations of the disturbance source, and to calculate the comprehensive source tracing confidence level.

[0016] The present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of a method for locating and tracing multiple disturbance sources in power quality of a power distribution network.

[0017] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of a method for locating and tracing multiple disturbance sources in power quality of a power distribution network.

[0018] The beneficial effects of this invention are: based on the observability analysis method, taking into account the topology and disturbance distribution characteristics of the distribution network, the location and number of monitoring points are optimized and determined, thereby realizing the optimized configuration of the monitoring network.

[0019] Multi-scale time-frequency analysis technology is used to extract comprehensive features of disturbances. Combined with hierarchical classification strategy and disturbance separation algorithm, the accuracy of disturbance identification is improved.

[0020] By integrating multi-point monitoring data and comprehensively utilizing distribution network topology information, the accuracy and reliability of positioning have been significantly improved. A comprehensive confidence assessment mechanism ensures the reliability of the positioning results, providing accurate positioning information for power quality management. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic flowchart of a method for locating and tracing multiple disturbance sources in power quality of a distribution network, provided as an embodiment of the present invention. Detailed Implementation

[0023] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0024] Example 1, referring to Figure 1This is the first embodiment of the present invention, which provides a method for locating and tracing multiple sources of power quality disturbances in a power distribution network, including: S1: Based on the distribution network topology, line impedance parameters, and historical disturbance distribution, an optimal monitoring point configuration model is established through observability analysis to determine the optimal monitoring point location and number.

[0025] S2: Deploy power quality monitoring devices at the optimal monitoring point to collect three-phase voltage and current waveform data, and perform outlier detection and signal denoising to obtain a pre-processed clean signal.

[0026] S3: Extract time-domain, frequency-domain, and time-frequency-domain features from the preprocessed clean signal to construct a comprehensive feature vector.

[0027] S4: Based on the comprehensive feature vector, a hierarchical classification strategy is adopted to identify the type of disturbance, and disturbance separation is performed in the scenario of multiple disturbances coexisting. The data fusion method is used to locate the source of the disturbance.

[0028] S5: Based on the location of the disturbance source, establish a disturbance propagation model, analyze the propagation path and impact range of the disturbance in the distribution network, and generate a disturbance source tracing report.

[0029] It should be noted that, based on multi-point power quality monitoring data from the distribution network, this method achieves accurate location and tracing of multiple power quality disturbance sources by optimizing monitoring point configuration, extracting multi-scale features, classifying and identifying disturbances, fusing and merging data at multiple points for localization, and analyzing propagation paths. First, based on the distribution network topology and historical disturbance data, the optimal monitoring point locations are determined using observability analysis methods. Then, the voltage and current waveform data collected from each monitoring point are preprocessed and multi-scale features are extracted, and machine learning algorithms are used to automatically identify disturbance types. Next, by combining the disturbance characteristics and time information from each monitoring point, a data fusion algorithm is used to locate the disturbance source. Finally, a disturbance propagation model is established based on distribution network parameters to analyze the propagation path and impact range of the disturbance. The entire method realizes a complete technical chain from monitoring network design to disturbance source tracing.

[0030] Example 2 is an embodiment of the present invention. Based on the above embodiment, a method for locating and tracing multiple disturbance sources of power quality in a power distribution network is provided.

[0031] Furthermore, in this embodiment of the application, step S1 establishes an optimal configuration model for monitoring points based on the distribution network topology, line impedance parameters, and historical disturbance distribution through observability analysis, and determines the optimal location and number of monitoring points. Specific steps include: The system first establishes the node admittance matrix based on the distribution network topology. The distribution network contains n nodes, and the admittance matrix is... Define the monitoring point configuration vector. ,in This indicates whether node i has a monitoring device installed. Indicates installation. This indicates that it is not installed.

[0032] The observability of a monitoring network is defined as: In the formula, O represents the overall observability. Let be the observability of node i. The formula for calculating the observability of node i is: In the formula, M is the set of monitoring points. Let be the electrical distance (in ohms) from node i to monitoring point j. This is the distance attenuation coefficient, with a value of [value missing]. The electrical distance is calculated using the nodal admittance matrix: In the formula, Nodal impedance matrix Element.

[0033] The optimization objective of monitoring point configuration is to minimize the number of monitoring devices while meeting observability requirements. In the formula, To meet the minimum observability requirement, according to the power quality monitoring specifications for distribution networks, the minimum observability is taken as follows: =0.85. A genetic algorithm is used to solve this integer programming problem, with a population size of 50, 100 generations, a crossover probability of 0.8, and a mutation probability of 0.1.

[0034] Optimized monitoring point configuration scheme Provide guidance on the on-site deployment of monitoring devices.

[0035] Furthermore, in this embodiment, step S2 involves deploying a power quality monitoring device at the optimal monitoring point to collect three-phase voltage and current waveform data, and performing outlier detection and signal denoising to obtain a pre-processed clean signal. The specific steps include: Based on the location of the monitoring points, power quality monitoring devices are deployed at each monitoring point to collect three-phase voltage and current waveform data. The sampling frequency is set to... It can capture harmonic components up to the 50th order. The raw data collected at each monitoring point is as follows: In the formula, For the original dataset of the k-th monitoring point, , , The instantaneous values ​​of the three-phase voltages A, B, and C are given. , , This represents the instantaneous value of the three-phase current.

[0036] The collected data is preprocessed, starting with the detection of outliers. Anomaly detection metrics are defined as follows: In the formula, For voltage deviation, Let be the instantaneous voltage value at time t. This is the statistical average of the voltage. When... It was determined to be an outlier at that time. This represents the voltage standard deviation. Outliers were corrected using cubic spline interpolation.

[0037] The signal is denoised using a wavelet thresholding method. A 5-level decomposition is performed using the db4 wavelet basis, and a soft thresholding function is employed. In the formula, These are the denoised wavelet coefficients. These are the original wavelet coefficients. The threshold is a general threshold. , Here, represents the noise standard deviation estimate, and N is the signal length. The noise standard deviation is estimated using the highest-level detail coefficients. In the formula, represents the detail coefficient of the highest decomposition layer, and median represents the median.

[0038] In an optional embodiment, outlier detection can also employ a moving window of fixed time length. For the signal data within the window, its local mean and standard deviation are calculated in real time. The current data point is compared with the local mean of the window; if the deviation exceeds a threshold (e.g., 2.5 times) set based on the standard deviation of that window, it is determined to be a local outlier.

[0039] In an optional embodiment, outlier detection can also estimate a normal voltage change rate range using historical normal data. During real-time monitoring, if the change rate of a sampling point from the previous point significantly exceeds this normal range, that point is determined to be an abrupt anomaly caused by interference.

[0040] Furthermore, in this embodiment, step S3 extracts time-domain, frequency-domain, and time-frequency-domain features from the preprocessed clean signal to construct a comprehensive feature vector. Specific steps include: Multi-scale feature extraction is performed on the preprocessed voltage and current signals to comprehensively characterize the disturbance properties. First, time-domain features are extracted, and the effective voltage value sequence is calculated: In the formula, W represents the effective voltage value at the k-th monitoring point during the n-th time window, where W is the window length (corresponding to one power frequency cycle). ), The window sliding interval (take half a cycle) The sampling period is specified. Voltage sags, swells, and interruptions are detected using a sequence of effective values. In the formula, This is a voltage sag event identifier. The rated voltage is 10000V for a 10kV distribution network.

[0041] Then, frequency domain features are extracted, and harmonic components are analyzed using Fast Fourier Transform (FFT): In the formula, The h-th harmonic content is The effective value of the h-th harmonic is... This is the fundamental effective value. Calculate the total harmonic distortion (THD): In the formula, THD is the total harmonic distortion rate, and H is the highest harmonic order considered (taken as 50th).

[0042] To identify interharmonics and time-varying disturbances, the S-transform is used for time-frequency analysis. The S-transform is defined as: In the formula, The result of the S-transform. For signal, For Gaussian window functions: In the formula, the window width is adaptively adjusted with frequency. Time-frequency energy features are extracted from the S-transform results: In the formula, Let be the time-frequency energy at time t. This feature is used to identify impulsive disturbances and oscillatory transients.

[0043] Construct feature vectors that integrate time-domain, frequency-domain, and time-frequency-domain features: In the formula, Let k be the feature vector of the kth monitoring point. For the statistical characteristics of the effective voltage value, Total harmonic distortion (THD) For characteristic harmonic content, For maximum time-frequency energy, is the time-frequency energy change rate.

[0044] In an optional embodiment, frequency domain feature extraction can also involve multiplying the current or voltage signal by a window function and performing an FFT on the windowed signal. Using the amplitudes of several spectral lines near the peak frequency, an interpolation algorithm is used to correct the frequency, amplitude, and phase of the harmonics to obtain harmonic parameters, which are then used to calculate the harmonic content and THD.

[0045] In an optional embodiment, frequency domain feature extraction can also be achieved by resampling and digital filtering to "amplify" the spectrum of the frequency band of interest across the entire analysis band. This frequency-shifted and filtered signal is then subjected to FFT analysis.

[0046] Furthermore, in this embodiment, step S4, based on the comprehensive feature vector, employs a hierarchical classification strategy to identify the disturbance type, separates disturbances in scenarios with multiple coexisting disturbances, and uses a data fusion method to locate the disturbance source. Specific steps include: Based on the extracted feature vectors, a hierarchical classification strategy is used to identify the type of disturbance. First, a first-layer classifier is established, classifying disturbances into two main categories: steady-state disturbances and transient disturbances. The discriminant function is: In the formula, This is the result of the first-level classification. The threshold for the time-frequency energy change rate is determined through statistical analysis of training samples. .

[0047] For steady-state disturbances, a second-layer classifier is established to further subdivide them into types such as harmonics, voltage deviations, and three-phase imbalances. A Support Vector Machine (SVM) classifier is used, with the Radial Basis Function (RBF) chosen as the kernel function. In the formula, For kernel function, For kernel parameters, take d is the dimension of the feature vector. The SVM classification decision function is: In the formula, For Lagrange multipliers, Here, b represents the training sample labels, and b is the bias term. This represents the number of support vectors. The steady-state perturbation categories are obtained using a multi-class SVM. .

[0048] Transient disturbances are classified based on their duration and energy distribution characteristics. The transient duration is defined as follows: In the formula, The duration of the disturbance. and These are the start and end times of the disturbance, detected by... The time periods exceeding the threshold are identified. They are categorized based on duration and maximum energy amplitude: In the formula, For transient disturbances, The power frequency period is 0.02s. This is the lowest voltage during the disturbance. The dominant frequency is determined by the frequency at the point of maximum energy in the S-transform result.

[0049] For scenarios with multiple perturbations, a perturbation separation algorithm is employed. Harmonic and transient components are separated based on morphological filtering. In the formula, Harmonic components, It is a transient component. This represents the convolution operation. It is a low-pass morphological filter with a cutoff frequency of 2.5kHz. The separated components are then subjected to type identification.

[0050] By combining the disturbance identification results and feature information from various monitoring points, a multi-point data fusion method is used to locate the disturbance source. Time synchronization verification is performed to ensure that all monitoring points observe the same disturbance event. Time synchronization criteria are defined as follows: In the formula, The disturbance detection time difference between monitoring points k and l. and For each time stamp where a disturbance was detected To determine the maximum allowable time difference, a sampling period is selected. Monitoring data points that meet the time synchronization criteria are grouped into the same event.

[0051] In an optional embodiment, time synchronization verification can also be based on absolute time reference synchronization. Specifically, each power quality monitoring device is equipped with a high-precision time synchronization module, unifying all monitoring points to Coordinated Universal Time (UTC) as the absolute time reference. When a disturbance occurs, each monitoring point records data with a high-precision absolute timestamp. During data fusion, the absolute times of the disturbance recorded by each point are directly compared. If the difference is within the allowable error range (e.g., less than 1 millisecond), it is determined to be a synchronization event.

[0052] In an optional embodiment, time synchronization verification can also involve injecting a high-frequency reference signal with extremely low amplitude and known characteristics at the point of common coupling of the distribution network. All monitoring points continuously monitor this characteristic signal. When a disturbance occurs in the system and location is required, the time consistency of the data at each monitoring point is calibrated or verified by checking the phase offset or arrival time of the reference signal recorded by each monitoring point, thereby completing the synchronization verification.

[0053] For locating harmonic disturbance sources, the harmonic power direction method is used. The active power of the h-th harmonic at monitoring point k is calculated as follows: In the formula, The active power of the h-th harmonic is and These are the effective values ​​of harmonic voltage and current. This represents the phase difference between harmonic voltage and current. The harmonic direction index is defined as follows: In the formula, For harmonic direction, This indicates that harmonics flow from the node to the line (this side is the harmonic source). This indicates that harmonics flow from the line to the node. By analyzing the harmonic direction at each monitoring point, the range of harmonic source locations can be narrowed down.

[0054] Further, the harmonic impedance method is used for precise location. The formula for calculating the harmonic impedance of the harmonic source node is: In the formula, For harmonic source impedance, The harmonic impedance of the line from monitoring point k to node s is calculated from the line parameters: In the formula, For line resistance, Here, h represents the line reactance, and h represents the harmonic order. For distribution networks... For each candidate node, calculate its harmonic impedance and select the node with the smallest positive impedance as the harmonic source location. In the formula, For locating the harmonic source node, Let Re be the set of candidate nodes, where Re represents taking the real part.

[0055] For locating transient disturbance sources, the traveling wave time difference method is used. The propagation speed of the disturbance on the line is: In the formula, v is the propagation speed, and c is the speed of light ( ), The relative permittivity of the line is 1.0 for overhead lines and 2.3 for cables. The location of the disturbance source is calculated based on the time difference between the arrival times of the disturbance at different monitoring points. In the formula, Let k be the distance from the disturbance source to the monitoring point k. Let l be the time difference between the arrival of the disturbance at monitoring points l and k. The line length between monitoring points k and l is given (obtained from the power distribution network GIS system). Using multiple pairs of monitoring points, the coordinates of the disturbance source are determined using the least squares method. In the formula, The distance is calculated based on the assumed location of the disturbance source. M represents the actual measured distance, and M represents the number of monitoring point pairs.

[0056] In an optional embodiment, harmonic disturbance source location can also utilize the distribution network topology and line parameters to establish a system harmonic model. Harmonic voltages and currents measured at multiple monitoring points are used as quantities, and a state estimation algorithm is employed to estimate the harmonic voltages of all nodes in the system. By analyzing the estimated harmonic voltage amplitudes at each node, it is generally assumed that the harmonic voltage amplitude at the node where the harmonic source is located will be significantly higher, thus locating the disturbance source.

[0057] In an optional embodiment, harmonic disturbance source location can also be achieved by noting that the equivalent harmonic impedance values ​​calculated from upstream and downstream monitoring points will change significantly within the line segment where the harmonic source is located. By comparing the amplitude and phase angle characteristics of the equivalent harmonic impedance at adjacent monitoring points, it can be determined between which two monitoring points the harmonic source is located, thereby achieving segment location.

[0058] Furthermore, in this embodiment of the application, step S5 establishes a disturbance propagation model based on the located disturbance source, analyzes the propagation path and impact range of the disturbance in the distribution network, and generates a disturbance source tracing report. Specific steps include: For harmonic disturbances, the harmonic power flow calculation method is used. The formula for calculating the h-th harmonic voltage distribution in the distribution network is: In the formula, The harmonic voltage vectors at each node. Let h be the impedance matrix of the h-th harmonic. Inject a current vector into the harmonic source. The harmonic impedance matrix is ​​calculated based on the line parameters: In the formula, Let h be the admittance matrix of the h-th harmonic, and its elements are: , and Let be the resistance and reactance of line ij.

[0059] The amplitude of the harmonic source injected current is inferred from the location results in step five and the measured values ​​at the monitoring points: In the formula, Inject current into the h-th harmonic of the harmonic source. The harmonic voltage measured at monitoring point k. Let k be the impedance from the harmonic source to monitoring point k. Calculate the harmonic voltage at each node of the distribution network.

[0060] Define the harmonic propagation attenuation coefficient: In the formula, This is the ratio of the harmonic voltage at a distance d from the harmonic source to the harmonic voltage at the harmonic source. Let be the harmonic voltage at the node at a distance d. Let be the harmonic voltage at the harmonic source node. Statistically analyze the attenuation coefficients at different distances and fit the harmonic attenuation law.

[0061] For transient disturbances, a traveling wave propagation model is adopted. The attenuation law of transient disturbances propagating on the line is as follows: In the formula, Let be the transient voltage amplitude at a distance d from the disturbance source. The transient voltage amplitude at the disturbance source. This refers to the attenuation coefficient. The attenuation coefficient is calculated based on the line parameters: In the formula, R is the resistance per unit length of the line. The characteristic impedance of the line. Let L and C be the inductance and capacitance per unit length of the line, respectively. Calculate the propagation amplitude of the disturbance along each line and identify the affected lines.

[0062] Criteria for defining the scope of disturbance impact: In the formula, For the set of affected nodes, and Let be the harmonic voltage and transient voltage at node i, respectively. To influence the threshold. For harmonics, take... (i.e., 5% voltage distortion); for transients, take... (i.e., a 10% voltage deviation).

[0063] Based on the foregoing analysis results, a complete disturbance source tracing report is generated. The report includes the disturbance source location, disturbance type, disturbance characteristic parameters, impact range, and mitigation recommendations. The comprehensive source tracing confidence level is defined as follows: In the formula, To trace the overall confidence level, To identify the confidence level for the type of disturbance, For location reliability, For propagation analysis confidence level, weighting coefficients = 0.4、 = 0.4、 = 0.2 is determined based on the degree of influence of each module on the final result.

[0064] Location reliability is calculated based on the consistency of multiple measurements: In the formula, The standard deviation of the distance to the disturbance source is calculated for different monitoring points. The standard deviation is the average distance. The smaller the standard deviation, the more consistent the positioning results and the higher the confidence level.

[0065] The confidence score for propagation analysis is calculated based on the degree of agreement between the model's predicted values ​​and the measured values. In the formula, The k-harmonic voltage at the monitoring point predicted by the propagation model These are actual measured values. This represents the number of monitoring points.

[0066] when When the source tracing results are reliable, it is determined that the results are reliable; when At times, manual review is required; when If the source tracing results are questionable, it is recommended to increase the number of monitoring points or extend the monitoring period.

[0067] Based on the type and severity of the disturbance, mitigation recommendations are provided. For harmonic disturbances, the filter capacity calculation formula is: In the formula, For the required reactive power capacity of the filter, For the h-th harmonic reactance, The maximum harmonic voltage before treatment, This refers to the harmonic voltage limit.

[0068] For voltage sag disturbances, assess whether a dynamic voltage restorer (DVR) is needed. The DVR capacity calculation formula is as follows: In the formula, For DVR capacity, This is the lowest voltage sag. For load current, The power factor.

[0069] The traceability report is stored in a database, supporting historical queries and statistical analysis, providing data support for power quality management of the distribution network.

[0070] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that: This embodiment also provides a power quality multi-disturbance source location and source tracing analysis system for distribution networks, including: a monitoring point configuration module, a data acquisition and preprocessing module, a feature extraction and recognition module, a location and analysis module, and a source tracing output module; The monitoring point configuration module is used to establish an optimal monitoring point configuration model based on the distribution network topology, line impedance parameters, and historical disturbance distribution, and to determine the optimal monitoring point location and number. The data acquisition and preprocessing module acquires three-phase voltage and current waveform data, performs outlier detection and signal denoising, and obtains a clean preprocessed signal. The feature extraction and recognition module is used to extract time-domain, frequency-domain, and time-frequency-domain features from the preprocessed clean signal, construct a comprehensive feature vector, and identify the disturbance type based on the comprehensive feature vector using a hierarchical classification strategy. The location and analysis module is used to integrate the disturbance identification results and time information of each monitoring point, use data fusion methods to locate the disturbance source, and establish a disturbance propagation model to analyze the propagation path and impact range of the disturbance in the power distribution network. The source tracing output module is used to generate a source tracing report, including the location, type, impact range, and governance recommendations of the disturbance source, and to calculate the comprehensive source tracing confidence level.

[0071] This embodiment also provides an electronic device, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the method for locating and tracing multiple disturbance sources of power quality in a distribution network as proposed in the above embodiment.

[0072] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements a method for locating and tracing multiple disturbance sources in power quality of a distribution network as proposed in the above embodiment.

[0073] The storage medium proposed in this embodiment and the method for locating and tracing multiple disturbance sources in power quality of a distribution network proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0074] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0075] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0076] Example 4 is an embodiment of the present invention, used to verify a method for locating and tracing multiple disturbance sources in power quality of a distribution network.

[0077] This simulation example is based on a typical 10kV distribution network in a city. The network includes one 110kV substation and five 10kV outgoing lines, with a total main line length of approximately 35 kilometers and a load capacity of approximately 50MVA. The network connects to three distributed photovoltaic power stations (total capacity 5MW), two electric arc furnace loads, and several nonlinear loads. A detailed model of the distribution network was built on the PSCAD / EMTDC platform, simulating two months of actual operation scenarios, including normal operation, harmonic pollution, voltage sags, and interharmonics. The simulation sampling frequency was set to 12.8kHz, generating approximately 450 million data points. Based on this method, observability analysis was used to determine the optimal monitoring point configuration scheme, and nine power quality monitoring devices were deployed in the distribution network. The positioning accuracy and source tracing accuracy of this method were verified by injecting disturbances such as harmonic sources and short-circuit faults at different locations. During the simulation, the traditional single-point positioning method and the multi-point fusion positioning method of this patent were compared to evaluate the improvement in positioning accuracy.

[0078] Table 1 Comparison of the effects of optimized configuration of monitoring points

[0079] This method achieves high observability (0.87) with a minimum number of monitoring points through observability optimization, reducing configuration costs by 10-40%. Positioning accuracy reaches 178 meters, significantly outperforming other methods. The false negative rate is only 4.2%, mainly occurring in minor disturbances in monitoring blind zones, resulting in optimal overall performance.

[0080] Table 2 Comparison of Disturbance Type Identification Accuracy

[0081] This method employs multi-scale feature extraction and hierarchical classification strategies, achieving an average recognition accuracy of 96.9% for seven common perturbation types, which is 12.3 percentage points higher than the traditional FFT method and 7.1 percentage points higher than the wavelet analysis method. In scenarios with multiple perturbations, the accuracy remains at 93.8% through a perturbation separation algorithm, validating the method's robustness.

[0082] Table 3. Verification of Disturbance Source Location Accuracy

[0083] Verification was conducted by injecting disturbance sources at six different locations. The average positioning error of this patented multi-point fusion positioning method was 130 meters, which is 78.6% higher than the single-point positioning method and 63.7% higher than the two-point positioning method. The average positioning time was 2.0 seconds, meeting real-time requirements. The average overall confidence level reached 0.86, indicating that the positioning results are reliable.

[0084] Table 4. Verification of Propagation Path Analysis

[0085] The propagation path analysis model accurately predicted the disturbance's impact range, with an average prediction error of only 5.3%. Through harmonic power flow calculations and traveling wave propagation models, the critical propagation paths and the furthest affected nodes were accurately identified, providing a scientific basis for mitigation solutions. The model can predict the disturbance's attenuation at different nodes, guiding the optimization of monitoring points and the setting of protection devices.

Claims

1. A method for locating and tracing multiple sources of power quality disturbance in a distribution network, characterized in that: include, Based on the distribution network topology, line impedance parameters, and historical disturbance distribution, an optimal monitoring point configuration model is established through observability analysis to determine the optimal monitoring point location and number. Power quality monitoring devices are deployed at the optimal monitoring points to collect three-phase voltage and current waveform data, and outlier detection and signal denoising are performed to obtain a pre-processed clean signal. Time-domain, frequency-domain, and time-frequency-domain feature extraction is performed on the preprocessed clean signal to construct a comprehensive feature vector; Based on the comprehensive feature vector, a hierarchical classification strategy is adopted to identify the type of disturbance, and disturbance separation is performed in the scenario of multiple disturbances coexisting. The data fusion method is used to locate the source of the disturbance. Based on the location of the disturbance source, a disturbance propagation model is established to analyze the propagation path and impact range of the disturbance in the distribution network and generate a disturbance source tracing report.

2. The method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in claim 1, characterized in that: The determination of the optimal location and number of monitoring points includes: using observability analysis methods to calculate the observability of each node based on the distribution network topology and line parameters; and using an optimization algorithm to minimize the number of monitoring devices, thereby obtaining an optimized monitoring point configuration scheme while meeting the observability threshold.

3. The method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in claim 2, characterized in that: The process of obtaining a preprocessed clean signal involves collecting three-phase voltage and current waveform data by deploying power quality monitoring devices at each monitoring point, and processing the raw data using anomaly detection algorithms and signal denoising techniques to obtain a clean signal for subsequent feature extraction.

4. The method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in claim 3, characterized in that: The construction of the comprehensive feature vector includes extracting time-domain features, frequency-domain features, and time-frequency-domain features. The hierarchical classification strategy first distinguishes between steady-state disturbances and transient disturbances, then further identifies specific disturbance types, and employs a disturbance separation algorithm to decompose composite disturbance signals in scenarios with multiple disturbances coexisting.

5. The method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in claim 4, characterized in that: The method of locating the disturbance source using data fusion includes performing time synchronization verification and ensuring that the same disturbance event is observed by comparing the disturbance detection time difference of each monitoring point. Define time synchronization criteria: In the formula, The disturbance detection time difference between monitoring points k and l. and The timestamps for the detection of disturbances (from step two). To determine the maximum allowable time difference, a sampling period is selected. Observational data from monitoring points that meet the time synchronization criteria are classified as the same event; For locating harmonic disturbance sources, the harmonic power direction method and harmonic impedance method are used. By calculating the harmonic active power and direction index at each monitoring point and combining it with the line harmonic impedance, the location of the harmonic source is determined. Calculate the active power of the h-th harmonic at monitoring point k: In the formula, The active power of the h-th harmonic is and These are the effective values ​​of harmonic voltage and current. For harmonic voltage and current phase; define harmonic direction index: In the formula, For harmonic direction, This indicates that harmonics flow from the node to the line. This indicates that harmonics flow from the line to the node; by analyzing the harmonic direction at each monitoring point, the range of harmonic source locations is narrowed down. For locating transient disturbance sources, the traveling wave time difference method is adopted. By calculating the time difference and propagation speed of the disturbance reaching different monitoring points, and combining the distribution network topology, the least squares method is used to solve for the coordinates of the disturbance source.

6. The method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in claim 5, characterized in that: The establishment of a disturbance propagation model to analyze the propagation path and impact range includes, for harmonic disturbances, using the harmonic power flow calculation method, calculating the harmonic voltage distribution at each node of the distribution network through the harmonic impedance matrix and the current injected by the harmonic source, and defining the harmonic propagation attenuation coefficient to analyze the attenuation law; For transient disturbances, a traveling wave propagation model is adopted. The attenuation law of transient disturbances propagating on the line is as follows: In the formula, Let be the transient voltage amplitude at a distance d from the disturbance source. The transient voltage amplitude at the disturbance source. The attenuation coefficient; The attenuation coefficient is calculated based on the line parameters: In the formula, R is the resistance per unit length of the line. The characteristic impedance of the line. L and C are the inductance and capacitance per unit length of the line, respectively; Calculate the propagation amplitude of the disturbance along each line and identify the affected lines; define the criteria for determining the scope of the disturbance's impact. In the formula, For the set of affected nodes, and Let be the harmonic voltage and transient voltage at node i, respectively. To influence the threshold.

7. The method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in claim 6, characterized in that: The generation of the disturbance source tracing report includes calculating the comprehensive tracing confidence score, which is obtained by weighting the disturbance type identification confidence score, location confidence score, and propagation analysis confidence score. Generate a source tracing report, including the location coordinates of the disturbance source, the type of disturbance, characteristic parameters, the scope of impact, and mitigation recommendations; Based on the type and degree of disturbance, suggestions are given for calculating the filter capacity or dynamic voltage restorer capacity.

8. A system for locating and tracing multiple sources of power quality disturbance in a distribution network, comprising the method for locating and tracing multiple sources of power quality disturbance in a distribution network as described in any one of claims 1 to 7, characterized in that, include: The system includes a monitoring point configuration module, a data acquisition and preprocessing module, a feature extraction and recognition module, a location and analysis module, and a source tracing output module. The monitoring point configuration module is used to establish an optimal monitoring point configuration model based on the distribution network topology, line impedance parameters, and historical disturbance distribution, and to determine the optimal monitoring point location and number. The data acquisition and preprocessing module acquires three-phase voltage and current waveform data, performs outlier detection and signal denoising, and obtains a preprocessed clean signal. The feature extraction and recognition module is used to extract time-domain, frequency-domain, and time-frequency-domain features from the preprocessed clean signal, construct a comprehensive feature vector, and identify the disturbance type based on the comprehensive feature vector using a hierarchical classification strategy. The positioning and analysis module is used to integrate the disturbance identification results and time information of each monitoring point, use data fusion methods to locate the disturbance source, and establish a disturbance propagation model to analyze the propagation path and impact range of the disturbance in the power distribution network. The source tracing output module is used to generate a disturbance source tracing report, including the location, type, impact range, and governance recommendations of the disturbance source, and to calculate the comprehensive source tracing confidence level.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for locating and tracing multiple disturbance sources in power quality of a distribution network as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for locating and tracing multiple disturbance sources of power quality in a power distribution network as described in any one of claims 1 to 7.

Images