Power Quality Disturbance Auto Recognition Based On Wavelet And Artificial Neural Network

A novel approach to automatically detect and classify various types of power quality dynamic disturbances is presented in this dissertation. The wavelet Daubechies 3 is used to decompose the signals containing dynamic disturbances. Using the Genetic Algorithm (GA) based on the neural network the pattern characters can be classified by the mode classifier. Computation results show that the proposed method has good performance in speed and accuracy. Referring to the theory of line longitudinal differential protection, in this dissertation a new method is brought out to locate the disturbances source. According to the simulation of the different disturbances using the Simulink toolbox of Matlab, it is tested that the new mode has high accuracy and efficiency. Zhang Zhiyuan(Power System and Its Automation) Directed by Prof. Li Gengyin...