Signal Processing-based Artificial Intelligence Approach for Power Quality Disturbance Identification

Power Quality (PQ) disturbance detection is thought to be a very significant service that many utilities provide for their commercial and industrial customers. PQ disturbances affect the load that is connected to the supply, which is troubling to the consumers. Detection and classification of the electrical problem are very difficult to find out which can cause PQ problems. In this paper, the key PQ issues such as voltage sag, voltage swell, voltage interruption, harmonics, and transient events have been tested. It has been demonstrated that a new approach may be used to identify, localize, and examine the probability of classifying distinct forms of PQ disturbances. The basic idea is to divide a disturbance signal into a transparent and comprehensive representation using DWT and ST. Many mathematical processes are utilized to extract features from these decomposed signals. The signal decomposition technique is integrated with the feed-forward neural network model to develop the power quality problem identifier (detection and classification). The simulation results show that the proposed method is effective. The proposed method is also feasible and promising for real-time applications.