LAUSR

In the modern era, the more usage of the non-linear load has increased the importance of power quality (PQ) monitoring. This paper purposes the novel algorithm based on Multivariate singular spectral analysis (MSSA), Wavelet Packet Decomposition (WPD) and 1-Dimensional Convolution Neural Network (1-D-CNN) for monitoring, mitigation, and classification of power quality disturbances (PQDs). Twelve types of synthetic and simulated single and multiple PQDs data are generated from MATLAB R2017b and Modified IEEE 13-bus system using wind energy penetration. In this research, MSSA and WPD are decomposed into four levels to extract the statistical features such as energy, entropy, standard deviation, root mean square, skewness, and kurtosis. The experimental results are well explained to compare the best-suited feature extraction technique in terms of feature extraction accuracy and computational complexity.  Optimally selected features are fed to a convolution neural network (CNN) based softmax classifier for classification of PQ disturbances. The proposed algorithm is also tested under no noise and 20 dB to 50 dB noisy environment. The performance of the proposed method is compared with recently published articles to justify the competency of this study. The results show that the proposed framework has obtained reliable highest classification accuracy.