The paper proposes a novel fuzzy pattern recognition system for power quality disturbances. It is a two-stage system in which a mulitersolution S-transform is used to generate a set of optimal feature vectors in the first stage. The multiresolution S-transform is based on a variable width analysis window, which changes with frequency according to a user-defined function. Thus, the resolution in time or the related resolution in frequency is a general function of the frequency and two parameters, which can be chosen according to signal characteristics. The multiresolution S-transform can be seen either as a phase-corrected version of the wavelet transform or a variable window short time Fourier transform that simultaneously localizes both real and imaginary spectra of the signal. The features obtained from S-transform analysis of the power quality disturbance signals are much more amenable for pattern recognition purposes unlike the currently available wavelet transform techniques. In stage two, a fuzzy logic-based pattern recognition system is used to classify the various disturbance waveforms generated due to power quality violations. The fuzzy approach is found to be very simple and classification accuracy is more than 98% in most cases of power quality disturbances.