Abstract
Improving power quality disturbance (PQD) detection and automatic classification has been a major concern ever since the emergence of sensitive non-linear devices. The role of distributed generation in a power system is the main source of PQDs. Short-term and long-term duration single and multiple complex PQDs are difficult to monitor and need higher accuracy and time. This paper presents the analysis of different and distinctive combinations of PQDs. Variety of single and multiple PQD samples are generated using Matlab environment conferring to IEEE STD 1159-2009. Such disturbance samples are accurately detected and analyzed from waveform patterns using multi resolution analysis based discrete wavelet transform. The generation of samples and detection lies in fact that it can allow the feature extraction process for the training/testing sample features for machine learning based automatic recognition of disturbance types.