A Hybrid Island Detection Method Based on the Combination of Wavelet Neural Network and SMS

Abstract

The invention discloses a hybrid island detection method based on the combination of wavelet neural network and SMS. The technical scheme of the method is: in the first stage, a voltage signal is obtained by detecting the voltage signal on the common coupling point in a period and performing wavelet transformation. The average value of the absolute value of the wavelet coefficients in the period is used as the input signal of the BP neural network, and finally the BP neural network is used for pattern recognition; the second stage introduces the phase disturbance in the SMS algorithm, so that whether the frequency of the common coupling point voltage exceeds Threshold, and through frequency detection to determine whether the islanding phenomenon really occurs, so as to take corresponding islanding protection measures when the grid-connected power generation system is in the islanding state. The invention greatly improves the reliability of the isolated island detection, and greatly reduces the influence of the detection method on the power quality of the system under the premise of ensuring the identification rate of the isolated island.

Description

technical field [0001] An island detection method, especially a hybrid island detection method based on the combination of wavelet neural network and SMS. Background technique [0002] With the increase of human society's demand for electric energy and the advancement of electric power industry technology, distributed generation (DG) technology has received extensive attention and in-depth research in countries all over the world. The DG system represented by photovoltaic power generation and wind power generation is not only clean and environmentally friendly, but also has renewable energy. Compared with traditional centralized power generation, it also has the characteristics of small investment and flexible power generation methods. However, after a large number of distributed generation is connected to the grid, a series of new problems have arisen, and the island effect is one of the serious problems. Therefore, island detection is a necessary function for distributed g...

AI Technical Summary

Problems solved by technology

Document 2 Invention Patent "Island Detection Method Based on the Introduction of Particle Swarm Algorithm and Phase Perturbation" (Notice No. CN104638671A), this invention uses the combination of particle swarm wavelet neural network and phase perturbation method, but the introduction of particle swarm increases the computational complexity. complexity and detection time, especially as the load quality factor Q f The increase of the traditional SMS phase disturbance detection speed will slow down and even lead to the defect of islanding detection failure

Images