CN-122020600-A - Power quality disturbance identification method and system by time-frequency joint analysis

CN122020600ACN 122020600 ACN122020600 ACN 122020600ACN-122020600-A

Abstract

The invention discloses a method and a system for identifying the disturbance of the power quality of time-frequency joint analysis, which belong to the technical field of monitoring and analysis of the power quality of a power system, wherein the method comprises the following steps of synchronously sampling a voltage signal or a current signal in a tested power system to obtain an original time domain sampling signal; the method comprises the steps of carrying out time-frequency joint analysis on a preprocessed signal to obtain corresponding instantaneous frequency information, calculating an instantaneous frequency track of a disturbance signal, carrying out second-order change analysis on the instantaneous frequency track to generate instantaneous frequency change curvature parameters, calculating energy time sequence change curvature based on evolution information of disturbance energy changing along with time, determining starting time and ending time of an electric energy quality disturbance event, constructing a feature set in a time range defined by the starting time and the ending time, identifying the electric energy quality disturbance, and outputting disturbance types and corresponding time position parameters. The invention can realize high-precision identification and positioning of the disturbance of the complex electric energy quality.

Inventors

Yin Jiaoyan
GUO HAIXIA
SONG YICHANG
CAO YUE
GUAN YUXIN
GAO YE

Assignees

山西农业大学

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. The power quality disturbance identification method based on time-frequency joint analysis is characterized by comprising the following steps of: synchronously sampling a voltage signal or a current signal in a tested power system to obtain an original time domain sampling signal; Preprocessing the original time domain sampled signal, wherein the preprocessing comprises performing adaptive fundamental wave notch filtering processing on fundamental wave components in the signal; performing time-frequency joint analysis on the preprocessed signals to obtain energy distribution of the preprocessed signals on a time-frequency plane and evolution information of disturbance energy which is derived from the energy distribution and changes along with time, and synchronously obtaining corresponding instantaneous frequency information; Calculating an instantaneous frequency track of a disturbance signal based on the instantaneous frequency information, and performing second-order change analysis on the instantaneous frequency track to generate an instantaneous frequency change curvature parameter for representing the disturbance frequency evolution form; Calculating the energy time sequence change curvature based on the evolution information of the disturbance energy changing along with time, and determining the starting time and the ending time of the power quality disturbance event based on the abrupt change characteristic of the energy time sequence change curvature; And constructing a feature set for representing the disturbance form based on the instantaneous frequency change curvature parameter in a time range defined by the starting moment and the ending moment, identifying the power quality disturbance according to the feature set, and outputting the disturbance type and the corresponding time position parameter.
2. The method for recognizing the power quality disturbance by using the time-frequency joint analysis according to claim 1, wherein the time-frequency joint analysis adopts a synchronous compression type time-frequency conversion mode; the synchronous compression process enables energy to be concentrated along the instantaneous frequency track direction through redistributing the energy in the initial time-frequency representation; in the synchronous compression process, the calculation of the instantaneous frequency is obtained according to the phase change rate of the time-frequency coefficient.
3. The method for identifying the power quality disturbance by the time-frequency joint analysis according to claim 1, wherein the instantaneous frequency change curvature parameter is obtained by calculating at least a second derivative of an instantaneous frequency track, wherein the first derivative is used for representing a frequency drift rate, and the second derivative is used for representing a degree of bending of a frequency change direction; When the transient frequency change curvature parameters are generated, firstly, time alignment is carried out on transient frequency tracks corresponding to different frequency components, amplitude normalization processing is carried out on each component according to the frequency change amplitude of each component in the disturbance duration, and then, the corresponding change curvature parameters are calculated based on the aligned and normalized transient frequency tracks; and, the sequence of the transient frequency change curvature parameter in the disturbance duration is statistically analyzed, and at least one or more of mean value, peak value, change range and standard deviation are extracted.
4. The method for power quality disturbance recognition by time-frequency joint analysis according to claim 1, wherein determining the start time and the end time based on the abrupt change feature of the curvature of the energy time sequence variation comprises: Generating an energy time sequence based on evolution information of the disturbance energy with time; Smoothing the energy time sequence; calculating the numerical change rate of the smoothed energy time sequence to obtain an energy change curvature sequence for representing the change curvature of the disturbance energy time sequence; Detecting local extremum points exceeding an adaptive threshold in the energy change curvature sequence as boundary candidate points, wherein the adaptive threshold is generated according to the energy fluctuation level of a background steady-state signal before disturbance occurs; and carrying out time continuity and persistence constraint on the boundary candidate points to determine the starting moment and the ending moment.
5. The method for power quality disturbance identification by time-frequency joint analysis according to claim 1, wherein constructing a feature set for characterizing a disturbance morphology based on the transient frequency variation curvature parameter comprises: Generating an initial feature vector based on the transient frequency change curvature parameter corresponding to each frequency component; processing the transient frequency change curvature parameters corresponding to each frequency component in the disturbance duration time to optimize the initial feature vector; and weighting and fusing the optimized initial feature vector according to the disturbance energy duty ratio or the frequency stability index obtained by the time-frequency joint analysis of each frequency component.
6. The method for power quality disturbance identification by time-frequency joint analysis according to claim 1, wherein the method further comprises: preprocessing three-phase voltage signals or current signals acquired synchronously with the voltage signals or current signals respectively, wherein the preprocessing comprises the steps of executing adaptive fundamental wave notch filtering processing on fundamental wave components in the signals so as to weaken the fundamental wave components; Respectively executing time-frequency joint analysis on the preprocessed three-phase signals, and calculating the instantaneous phase difference between the phases at the corresponding time-frequency positions; calculating a deviation index reflecting the consistency of the three-phase time-frequency phases based on the instantaneous phase difference, wherein the deviation index is used for representing the consistency degree of disturbance characteristics among different phases; When the deviation of the phase consistency is lower than or equal to a preset threshold value, judging that the identified power quality disturbance is systematic disturbance; And outputting the judging result of the systematic disturbance, the locality or the asymmetric disturbance in a correlated manner with the disturbance type and the time parameter.
7. The method for recognizing power quality disturbance according to claim 6, wherein the deviation index reflecting three-phase time-frequency phase consistency is obtained by performing a statistical analysis on instantaneous phases of the respective corresponding frequency components, and the statistical analysis includes at least one or more of a phase difference average value, a phase difference fluctuation range, and a phase difference stability index.
8. The method for power quality disturbance recognition by time-frequency joint analysis according to claim 1, wherein the method further comprises evaluating measurement uncertainty of a disturbance recognition result; The measurement uncertainty is comprehensively calculated at least based on sampling noise level, transformation stability of time-frequency joint analysis, fluctuation condition of the instantaneous frequency change curvature parameter and consistency level of each phase-to-phase instantaneous phase difference when three-phase signals are identified; And the measurement uncertainty is associated with the disturbance type and the time position parameter and output.
9. The method for power quality disturbance identification by time-frequency joint analysis according to claim 1, wherein the method further comprises: Mapping the outputted disturbance type and the time parameter into corresponding electric energy quality evaluation parameters; Wherein the power quality evaluation parameter at least comprises a voltage sag depth, a disturbance duration or a harmonic influence degree.
10. The system for identifying the power quality disturbance by using time-frequency joint analysis is applied to the method for identifying the power quality disturbance by using time-frequency joint analysis as claimed in any one of claims 1 to 9, and is characterized by comprising the following steps: the synchronous sampling module is used for synchronously sampling voltage signals or current signals in the power system; The preprocessing module is used for carrying out self-adaptive fundamental wave notch filtering processing on the sampling signals; the time-frequency analysis module is used for executing time-frequency joint analysis and calculating an instantaneous frequency track; The feature generation module is used for calculating instantaneous frequency change curvature parameters and energy time sequence change curvature and constructing a feature set for representing disturbance forms; And the disturbance identification module is used for identifying the power quality disturbance type according to the feature set and outputting the disturbance type and the corresponding time position parameter thereof.

Description

Power quality disturbance identification method and system by time-frequency joint analysis Technical Field The invention relates to the technical field of electric energy quality monitoring and analysis of an electric power system, in particular to an electric energy quality disturbance identification method and an electric energy quality disturbance identification system for time-frequency joint analysis. Background With the large-scale grid connection of renewable energy sources and the wide access of power electronic equipment, the electric energy quality disturbance events in modern power systems are frequent and have complex forms. The disturbance events such as voltage sag, sag rise, harmonic wave, inter-harmonic wave, oscillation and the like generally have transient, non-stationarity and multi-component superposition characteristics, and are an important technical basis for ensuring safe and stable operation of a power grid and high-quality power supply of a user. Currently, the recognition method of the power quality disturbance is mostly based on time-frequency analysis technology, such as short-time fourier transform and wavelet transform. The method generally extracts energy, amplitude or statistical characteristics in a specific frequency band by carrying out time-frequency decomposition on signals, carries out disturbance type identification by combining threshold judgment or a classification method based on the statistical characteristics, and also introduces a data-driven classification model in part of schemes to carry out auxiliary judgment. However, the above method still has a certain limitation in practical application. On the one hand, the traditional time-frequency analysis method is limited by the inherent trade-off relation between the time domain resolution and the frequency domain resolution, and often appears as time-frequency representation energy diffusion and outline blurring in engineering application, and especially under the condition that disturbance frequency is rapidly changed or multiple components are tightly overlapped, the instantaneous frequency track of each disturbance component is difficult to accurately describe. On the other hand, the features extracted by the existing method are concentrated on low-order static features such as amplitude, energy or fixed frequency band power, and an effective quantitative description means is lacking for a dynamic process of energy and frequency change along with time in a disturbance generating process, so that the change rate and a trend inflection point thereof in a disturbance evolution process are difficult to reflect. Under the scene of obvious frequency dynamic change, disturbance composite superposition or disturbance boundary blurring, the recognition accuracy is easy to be reduced, and the problems of misjudgment of disturbance type, inaccurate positioning of start and stop moments, difficulty in effective separation of composite disturbance and the like occur. Disclosure of Invention This section is intended to outline some aspects of embodiments of the application and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description of the application and in the title of the application, which may not be used to limit the scope of the application. In order to solve the technical problems, the invention provides a method for identifying the disturbance of the power quality by using time-frequency joint analysis, which comprises the following steps: synchronously sampling a voltage signal or a current signal in a tested power system to obtain an original time domain sampling signal; Preprocessing the original time domain sampled signal, wherein the preprocessing comprises performing adaptive fundamental wave notch filtering processing on fundamental wave components in the signal; performing time-frequency joint analysis on the preprocessed signals to obtain energy distribution of the preprocessed signals on a time-frequency plane and evolution information of disturbance energy which is derived from the energy distribution and changes along with time, and synchronously obtaining corresponding instantaneous frequency information; Calculating an instantaneous frequency track of a disturbance signal based on the instantaneous frequency information, and performing second-order change analysis on the instantaneous frequency track to generate an instantaneous frequency change curvature parameter for representing the disturbance frequency evolution form; Calculating the energy time sequence change curvature based on the evolution information of the disturbance energy changing along with time, and determining the starting time and the ending time of the power quality disturbance event based on the abrupt change characteristic of the energy time sequence change curvature; And constructing a feature set for representing the disturbance form based on the