Detection and Classification of Faults in AC Microgrids based on Wavelet Transform

The incorporation of renewable energy sources and the increasing intricacy of microgrid systems are intertwined developments underscore the need for robust fault detection mechanisms to ensure reliable and stable operation. This research focuses on the application of wavelet transform for fault detection in a 5-bus microgrid system. The wavelet transform, recognized for its capacity to examine signals in both temporal and frequency domains, is employed to capture transient behaviors indicative of faults. Transmission lines are crucial components of power systems responsible for the efficient and reliable transfer of electricity over long distances. Yet their seamless operation can be hindered by various faults and disruptions. This paper proposes a novel approach to identify faults for microgrid leveraging wavelet-based analysis. The scheme aims to enhance fault detection accuracy by harnessing the capabilities of wavelet transform to provide valuable insights into both temporal and frequency domains. The proposed scheme begins by collecting data from various sensors within the microgrid, capturing critical parameters such as voltage, current, and frequency. Raw data is often noisy and requires preprocessing to optimize signal quality. Wavelet transform is subsequently utilized to break down the signal into its constituent frequencies components across multiple scales. This decomposition enables the extraction of distinctive features that signify anomalies caused by faults or disturbances. In light of the foregoing, this work intends to investigate the use of wavelet-based fault detection in microgrids, highlighting its advantages, difficulties, and potential to improve power grid performance as a whole.