Energy feature extraction and visualization of voltage sags using wavelet packet analysis for enhanced power quality monitoring.

Power quality (PQ) disturbances, such as voltage sags, are significant issues that can lead to damage in electrical equipment and system downtime. Detecting and classifying these disturbances accurately is essential for maintaining reliable power systems. This paper introduces a novel approach to voltage sag analysis by employing wavelet packet analysis combined with energy-based feature extraction to enhance PQ monitoring. The study decomposes voltage sag signals into different frequency bands to extract key features for disturbance detection. We compare six commonly used mother wavelets (db1, db4, db10, dmey, sym5, and coif5) to identify the most suitable wavelet for voltage sag detection. The energy distribution curve analysis is used to evaluate the energy characteristics of each wavelet's decomposition, with a focus on identifying the most effective signal features for PQ monitoring. The paper presents a thorough error analysis and compares the energy values extracted by different wavelet functions to demonstrate the reliability and accuracy of the proposed method. The results show that wavelet packet analysis significantly improves the detection and classification of voltage sag disturbances, providing a robust and efficient tool for real-time PQ monitoring. This study contributes to the development of advanced PQ monitoring systems by offering a more precise and computationally efficient method for voltage sag analysis, ultimately helping to protect electrical systems from potential damage and reducing operational costs. Wavelet packet analysis is applied as a novel feature extraction method for voltage sag detection by offering improved time-frequency analysis over traditional methods. Energy features are extracted from wavelet packet coefficients for sag identification. This work utilizes wavelet packet analysis to extract energy-based features for voltage sag detection.