Voltage notch detection using fuzzy expert system

In this paper a new method for detection of power quality events is proposed. Firstly, Fourier and wavelet transforms are applied to extract the specification of the disturbance signal. A fuzzifier system is then used to assign the membership functions to the extracted power quality events. For detection and classification of the disturbances, an expert system especially tailored to detect voltage notch is developed. In order to evaluate the ability of the proposed method, a 6-pulse and 12-pulse inverter system containing voltage notch has been simulated using Matlab. The results showed the suitability of the method in detecting voltage notch in both cases. This approach can be applied to develop a system for detection and classification of other power quality events with the emphasis being placed on fuzzy expert system rules.