A Support Vector Machine and Particle Swarm Optimization-Based Method for Detecting Fault Events and Locations in The Power System

• Main Authors: Li-Yuan Hsu, 徐立遠
• Other Authors: Wen-Kung Chang
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/70014054934293216845

碩士 === 國立中正大學 === 電機工程研究所 === 101 === With the advance of technology, people need higher quality of electricity than before. The disturbance in the power system will make voltage and current deviation from the rating. It may result in serious damage or equipment malfunction. Therefore, the transmission line fault detection and identification are one of the important studies in the power system. Traditional power system faults are divided into balanced and unbalanced faults. If a fault occurs in the power system, rapid diagnosis of the fault location can restore the power supply in shortest time and reduce outage times, as well as losses. In this thesis, EMTP / ATP is adopted to establish the power system model with the support vector machine (SVM) classification techniques. To precisely identify the fault event and the fault location, SVM is combined with particle swarm optimization (PSO) to adjust the feature parameters. Results show that the proposed method can provide fast and accurate fault diagnosis in the power system.