Intelligent Techniques for the Analysis of Power Quality Data in Electrical Power Distribution System

Abstract

This chapter is about the intelligent techniques for the analysis of power quality problems in electrical power distribution system. The problems related with electrical power industry are becoming more widespread, complex, and diversified. The behaviour of power distribution systems can be monitored effectively using artificial intelligence techniques and methodologies. There is a need of understanding the power system operations from power utility perspectives and application of computational intelligence methods to solve the problems of the power industry. The real power quality (PQ) data is taken from a power utility in Victoria Australia. Principal Component Analysis Technique (PCAT) is used to reduce the large number of PQ data attributes of the power distribution system. After the pre-processing of PQ data using PCAT, intelligent computational techniques will be used for the analysis of power quality data. Neural network techniques will be employed to estimate the values of PQ parameters of the power distribution system. The Feed Forward Back Propagation (FFBP) neural network and Recurrent Neural Networks (RNN) are used for intelligent estimation of PQ data. The results obtained through these intelligent techniques are compared with the real data of power utility in Victoria, Australia for stability, reliability and enhanced power systems performance.