Digital System for Detection and Classification of Power Quality Disturbance

In recent years power quality (PQ) has become a major concern to both power utilities and power customers. How to extract features of disturbances from large number of power signals and how to recognize them automatically are important for further understanding and improving of power quality. The paper presents an identification scheme for online monitoring and identification of power quality and system disturbances caused by nonlinear loads. The wide proliferation distributed renewable energy and green power sources, and rapid changes in utility load types require affordable and robust on-line data acquisition and expert identification systems, especially for utilization grid power systems. In this work, we propose a digital system for detection and classification of power quality disturbance using wavelet transform and multi-class support vector machines. The proposed technique allows creating such expert systems with the extensible knowledge base, which can be used for identification of power distortion events.