Design of Intelligent Fault Detection and Location Algorithm for Power System Based on Wavelet Analysis

In power system, system faults usually reflect abrupt changes in voltage and current signals, and timely and accurate detection of these abrupt signals is a prerequisite for reliable operation of relay protection devices. The distribution line is the terminal line directly connected to the users, and the fault point can be quickly located after the distribution line fails, which can improve the reliability of power supply for users. WT (wavelet transformation) is a widely used signal processing method in recent years, which has time-frequency localization performance. The main purpose of this paper is to study the intelligent detection and location algorithm of power system faults, and give a diagnosis and location method of power system faults under the fusion of electrical information and switching information by WT and BN (Bayesian network). Based on WT theory, the electrical information of fault information is decomposed and reconstructed, and the fault features are extracted. The comprehensive feature information is input into the trained BN for reasoning, so that the fault diagnosis and location can be realized. The simulation results show that the algorithm effectively improves the calculation process and accuracy, and improves the integrity of fault signal detection and location based on the complementarity of information composed of electrical quantities and switching quantities.