Power quality problem classification based on Wavelet Transform and a Rule-Based method

This paper describes a Wavelet Transform and Rule-Based method for detection and classification of various events of power quality disturbances. In this model, wavelet Multi-Resolution Analysis (MRA) technique was used to decompose the signal into its various details and approximation signals, and unique features from the 1st, 4th, 7th and 8th level detail are obtained as criteria for classifying the type of disturbance occurred. These features and together with the duration of disturbance of occurrence obtained from 1st level of detail, they form the criteria for a Rule-Based software algorithm for detecting different kinds of power quality disturbances effectively. It is presented in this paper that the choice of sampling frequency is important since it affects the average energy profile of the details and eventually may cause error in detection of power quality disturbances. The model is tested by using MATLAB toolbox. The simulation produces satisfactory result in identifying the disturbance and proof that it is possible to use this model for power disturbance classification. Since the method can reduce the number of parameters needed in classification, less memory space and computing time are required for its implementation. Thus it stands up to be a suitable model to be used in real time implementation through a dsPIC-based embedded system.