Wavelet and Machine learning based approach for Fault classification in AC Micro-grid system

Most common faults such as Line-ground (LG), Line-line-ground faults(LLG), Line-line-line-ground (LLLG), and Line-line(LL), Line-line-line(LLL) can not be detected by conventional protection relay in the micro-grid system as these faults are not detectable when the fault impedance is high. In the literature, many techniques have been proposed to overcome these challenges. However, these methods could not classify the type of faults and diagnose their severity. In this paper, a protection method is proposed for AC microgrids with Discrete wavelet transform (DWT) and Decision tree(DT) techniques to perform the job of fault classification. Testing of the proposed scheme has been performed against various faults like LG, LLG, LLLG, and LLL scenarios. The accuracy is validated on a test micro-grid system which is simulated using Power system computer-aided design (PSCAD) software.