An automatic voltage disturbance classification system based on Clonal Selection Algorithm

Classification of voltage disturbances in power systems is essential for modern society and can be very demanding according to the used method or the aimed accuracy. This paper presents a new intelligent approach aimed to automatically analyse power quality disturbances including sag, swell, outage, harmonics and normal waveform. The approach is based on Artificial Immune System and focuses on the application of a Clonal Selection Algorithm to extract features from disturbance waveforms and classify the disturbances in each 0.5 cycle of the fundamental frequency. Other important feature of the proposed approach is that it can be embedded since the resulted on-line classification tool achieves very low computational complexity. Comparisons and experimental results obtained from the application of the proposed method validate the approach and achieved a classification accuracy at least better than previous work.