Rule based system for power quality disturbance classification incorporating S-transform features

Detection and classification of power quality (PQ) disturbances in real-time is an important consideration to electric utilities and many industrial customers so that diagnosis and mitigation of such disturbances can be implemented quickly. This paper presents the design and development of a rule based system for intelligent classification of PQ disturbances using the S-transform features. A hardware system has been designed using advanced digital signal processor to provide fast data capture and processing of signals using the S-transform analysis. Distinct features of various disturbances are extracted from the S-transform analysis in which these features are used to formulate rules. A rule-based expert system is developed to automate the process of classifying the various types of disturbances. The disturbance classification results prove that the developed rule based system is more accurate than the neural network in classifying PQ disturbances such as voltage sag, swell, impulsive transient, notching and interruption.