This paper presents a method for detection and classification of power quality (PQ) disturbances using Stock-well's transform. PQ disturbances are generated using MATLAB as per IEEE-1159 standard. Various features of signals are extracted from the multi-resolution analysis based on Stockwell's transform. These features are used to classify PQ disturbances using the decision tree initialized Fuzzy C-means clustering. It is observed that the Fuzzy C-means clustering based classification yields satisfactory accuracy even under noisy conditions. The investigated PQ disturbances include voltage sag, swell, interruption, harmonics, notch, flicker, oscillatory transient, impulsive transient and spike. Effectiveness of proposed algorithm has been established by satisfactory results of various case studies. © 2016 Asian Institute of Technology.