CN-121456263-B - Sectional self-adaptive Fourier transform analysis method for electric energy quality disturbance

Abstract

The application discloses a segmented self-adaptive Fourier transform analysis method for power quality disturbance. The method comprises the steps of firstly collecting power quality disturbance signals in a power grid, carrying out Fourier transformation, obtaining peak envelope points in a frequency spectrum through a maximum envelope method to dynamically select characteristic frequency points and calculate frequency spectrum segmentation points, then optimizing Gaussian window parameters for each segment by adopting a composite objective function, and finally carrying out time-frequency analysis on the signals by utilizing the optimized window functions to obtain a time-frequency result with self-adaptive resolution. According to the method, the higher time resolution is allocated to the strong time-varying component through the segmentation self-adaptive resolution and the composite objective function optimization strategy, the higher frequency resolution is allocated to the near-steady-state component, the problems of spectrum overlapping and energy diffusion which are easy to occur in the traditional method are effectively avoided, the accuracy and the robustness of disturbance feature extraction are improved, and the method can be widely applied to power grid fault diagnosis, disturbance identification and early warning.

Inventors

• LI JIANMIN
• ZHANG JINLONG
• XIE WEI
• HUANG JIE
• Wen Qidi
• CAO DONG
• Mi Chengdong
• TANG QIANG
• LIN HAIJUN

Assignees

• 湖南师范大学

Dates

Publication Date

20260508

Application Date

20260105

Claims (5)

1. 1. The method for analyzing the segmented self-adaptive Fourier transform of the power quality disturbance signal is characterized by comprising the following steps of: S101, collecting PQDs signals in the power system to obtain a discrete signal sequence with the length of N ; S102, the discrete signal sequences are processed Performing fast Fourier transform FFT to obtain frequency spectrum , wherein, Is a frequency index; s103, obtaining a frequency spectrum by adopting a maximum envelope method Is satisfied by (1) Wherein the peak envelope point of the (c) is, among other things, For the peak envelope to be a peak value, , For the number of peak envelope points selected, Is an envelope threshold; s104, selecting characteristic frequency points , , wherein, The number of the characteristic frequency points; s105, according to the characteristic frequency point Calculating spectral segmentation points And : ; S106, adopting a composite objective function for each segment Optimizing corresponding Gaussian window parameters, wherein the composite objective function integrates the energy concentration measure E CM and 90% peak time width The expression of the composite objective function is as follows: , Wherein, the The value of the weight parameter is ; And S107, performing short-time Fourier analysis on the signal by using the optimized window function to obtain a time-frequency analysis result with self-adaptive time-frequency resolution.
2. 2. The method according to claim 1, wherein the envelope threshold is set in step S103 The amplitude of the fundamental wave is set to be 1-5%.
3. 3. The method according to claim 1, wherein the energy concentration measure E CM in step S106 is calculated as: , Wherein, the The normalization result of the short-time Fourier transform matrix mode is specifically as follows: 。
4. 4. the method according to claim 1, wherein in step S106 The calculation method comprises recording time sample corresponding to 10% amplitude increment between baseline value and peak value as And And calculate = 。
5. 5. The method according to claim 1, characterized in that the gaussian window parameter optimization in step S106 satisfies: , the constraint conditions are as follows: , Wherein, the For the standard deviation factor of each segment, In order to achieve a frequency resolution of the device, Is the minimum distance to the adjacent segmentation point.

Description

Sectional self-adaptive Fourier transform analysis method for electric energy quality disturbance Technical Field The application relates to the technical field of electric energy quality analysis of an electric power system, in particular to a segmented self-adaptive Fourier transform analysis method for electric energy quality disturbance. Background With the large-scale access of renewable energy sources and the wide application of power electronic equipment, the types of power quality disturbance and the combination forms thereof in the power system are increasingly complex. The accurate analysis of the disturbance has important significance for equipment fault diagnosis, evaluation and operation optimization. Traditional power quality analysis methods mainly rely on Short-time fourier Transform (STFT), wavelet Transform (Wavelet Transform, WT), hilbert-Huang Transform (HHT) and other time-frequency analysis techniques. The STFT is still the most widely applied time-frequency tool in engineering practice due to the characteristics of simple concept and high calculation efficiency. The basic idea of STFT is to localize the signal by a fixed window function, but the fixed window width results in its invariable time-frequency resolution, which is prone to spectral overlap or energy spread when analyzing PQDs where a strongly time-varying signal is mixed with a steady-state signal. To further increase the adaptability of STFT, researchers have proposed various improvements including generalized Short time Fourier transforms (Generalized Short-Time Fourier Transform, GSTFT), multi-Resolution Short time Fourier transforms (Multi-Resolution Short-Time Fourier Transform, MRSTFT), and the like. The methods control the window function by introducing a plurality of adjustable parameters, thereby enhancing the flexibility of the methods or adapting the methods to different application scenes. However, these methods still have some limitations in practical applications, as follows: 1. the standard STFT adopts a fixed window width, can not meet the time and frequency analysis requirements of time-varying signals at the same time, and is easy to generate a frequency spectrum overlapping phenomenon. 2. GSSTFT although window function optimization is introduced, adaptive adjustment capability for different disturbance types is still lacking, and complex disturbance types in a modern power system are difficult to effectively cope with. 3. MRSTFT depending on fixed frequency division, the lack of a mechanism for adaptively adjusting resolution according to the real-time characteristics of the signal has poor effect when processing signals with strong time-varying components and near-steady-state components. Under the background, an improved STFT analysis method with high precision, high robustness and high calculation efficiency is urgently needed, so that time-frequency resolution configuration is dynamically adjusted according to signal characteristics on the premise of not depending on predefined frequency band division, spectrum overlapping is avoided, disturbance positioning precision is improved, and analysis performance under a complex scene is improved. Disclosure of Invention The application aims to solve the technical problems of poor adaptability, serious spectrum overlapping, insufficient positioning accuracy and the like of the conventional PQDs time-frequency analysis method, and provides a segmented self-adaptive Fourier transform (GASTFT) method. According to the method, a segmented region can be dynamically determined according to the spectrum structure of PQDs signals without depending on a predefined frequency band, and adaptive optimization is carried out on each segmented window function parameter, so that a strong time-varying component obtains higher time resolution, and a steady-state or near-steady-state component obtains higher frequency resolution, and therefore the accuracy and the robustness of disturbance detection are remarkably improved. The method comprises the following steps: S101, collecting PQDs signals in the power system to obtain a discrete signal sequence with the length of N ; S102, the discrete signal sequences are processedPerforming fast Fourier transform (Fast Fourier Transform, FFT) to obtain frequency spectrum, wherein,Is a frequency index; s103, obtaining a frequency spectrum by adopting a maximum envelope method Is satisfied by (1); Wherein the peak envelope point of the (c) is, among other things,For the peak envelope to be a peak value,,For the number of peak envelope points selected,Is an envelope threshold; s104, selecting characteristic frequency points , wherein,The number of the characteristic frequency points; s105, according to the characteristic frequency point Calculating spectral segmentation pointsAnd:; S106, adopting a composite objective function for each segment) Optimizing the corresponding Gaussian window parameters, wherein the composite objective functi