application of fuzzy particle swarm optimization in detection and classification of single and combined power quality disturbances

detection & classification of power quality (pq) disturbances are the most important problems in distribution systems. in this paper, a new approach for the detection and classification of single and combined pq disturbances is proposed which utilizes fuzzy logic and particle swarm optimization (pso) algorithms. in this approach, first suitable features of the waveform of pq disturbances are extracted. extraction of these specifications is done based on the fourier and wavelet transforms. then, the proposed fuzzy systems make decision about the type of each of the pq disturbances by employing these specifications. the pso algorithm is used for accurate determination of each parameter of the membership functions of the systems. to test the proposed approach, the waveform of pq disturbances was assumed to be in a sampled form from single and combined categories. impulse, interruption, swell, sag, notch, transient, harmonic, and flicker phenomena are considered as single disturbances for voltage signal. more over, harmonic with swell, swell with harmonic, swell with transient, harmonic with sag, sag with harmonic and sag with transient are considered as combined disturbances for the voltage signal. simulation results denote capability of the proposed approach for identification of single and combined disturbances with about 99% accuracy.