Detection and classification of power quality disturbances based on Hilbert-Huang transform and feed forward neural networks

Abstract

This paper presents a hybrid detection method and classification Technique based on Hilbert-Huang Transform (HHT) and Feed Forward Neural Networks (FFNNs) to improve the efficient delivery and ensure accurate detection of quality disturbances in the electrical power grids. First, quantities characteristics of power quality disturbances (PQDs) are introduced according its parametrical conditions. Thereafter, a detection and recognition algorithm is used for single and multiple disturbances. Then, a decomposition process and features extraction using Empirical Mode Decompensation (EMD) is conducted for each of these distorted waveforms into Intrinsic Mode Functions (IMFs). Finally, these features are constructed using signal amplitude and frequency and then after fed to one of the powerful Artificial Intelligence Techniques in this field for training, evaluating and testing using (FFNNs) classifier to verify and confirm the effectiveness of the detection methodology.