Real and reactive power flow allocation in deregulated power system utilizing genetic-support vector machine technique

Abstract

This paper presents a technique to allocate the real and reactive power flow in deregulated power system environment by incorporating the hybridization of Genetic Algorithm and Least Squares Support Vector Machine (Genetic-SVM). The idea is to use GA to find the optimal values of hyper-parameters of LS-SVM and adapt a supervised learning approach to train the LS-SVM model. The manipulation of proportional sharing method (PSM) is utilized as a teacher. Based on converged load flow and followed by PSM for power flow allocation procedures, the description of inputs and outputs of the training data are created. The Genetic-SVM model will learn to identify which generators are supplying to which loads. In addition, the equivalent transmission model will be discussed in reactive power tracing methodology together with the concept of virtual load for both real and reactive power tracing methods. In this paper, 5- bus system and 25-bus equivalent system of southern Malaysia are used to show the effectiveness of the proposed method. The comparison with other method is also given.