A data mining method for obtaining global power quality index

The new development in power system such as restructuring and competetive electricity market make power quality an important factor in competition. However, to find a measure for power quality evaluation is very difficult due to many indices involved in power quality standards. For this reason, obtaining a single quantitative index based on the standard measurements has been a new challenge in recent researches. Data mining methods are required for this purpose due to the large amount of data obtained from power quality measurements. In this paper, a data mining method is proposed to determine a global index for power quality. The continuous and discrete indices of power quality are considered and the Unified Power Quality Index (UPQI) is presented for each power quality index, based on the method of incorporation and normalization. The indices normalized and classified. The power quality level of each distribution site is determined by the Fast Independent Component Analysis (Fast-ICA) algorithm. The power quality measurements of 313 distribution sites in Iran are used to classify the indices for different type of loads in the distribution system.