Combining Fuzzy Logic and Discrete Wavelet Transform for Accurate Fault and Inrush Current Classification in High Voltage Capacitor Banks

Abstract

The high voltage capacitor bank is a critical component in substations, essential for maintaining power quality and system stability. However, these banks are susceptible to faults and inrush currents, posing significant operational challenges. This paper presents a method for accurately classifying fault and inrush currents in high voltage capacitor banks using Fuzzy Logic and Discrete Wavelet Transform (DWT). The DWT decomposes current waveforms into frequency components, enabling the extraction of features that characterize faults and inrush currents. These features are processed by a Fuzzy Inference System (FIS), which classifies the events based on predefined rules and membership functions. The integration of DWT and FIS provides a robust framework for distinguishing between different types of faults and inrush currents with high accuracy. Simulation results demonstrate the proposed method’s efficacy, showing improved performance in classification accuracy and noise robustness compared to traditional techniques. This research enhances monitoring and protection systems in power networks, ensuring more reliable operation of high voltage capacitor banks. Implementing this method allows for better fault management and minimized downtime, leading to improved overall system efficiency and stability.