Skip to Main Content
This paper presents the concepts and application details of a new adaptive neuro-fuzzy intelligent tool for power quality analysis and diagnosis. The various conceptual details are stated and the application of such concepts to two test systems is illustrated. The work introduces a novel approach to power quality from a single system's perspective. For a given system, classification of normal from abnormal operation, as well as full abnormality diagnosis are performed. Adaptive fuzzy-based self-learning techniques are a key ingredient of the new approach. The validation of the new technique is accomplished by diagnosing the operational conditions of a three-phase induction motor and a three-phase rectifier bridge. The work paves the way toward an ultimate objective of developing an intelligent power quality diagnosis tool capable of predicting abnormal operation of individual power systems.