Application of Fuzzy Logic and Particle Swarm Optimization for Reactive Power Compensation of Radial Distribution Systems

The objective of this paper is to study optimal location and size of the capacitor to be placed in radial distribution feeders to improve the voltage profile and to reduce the energy loss. The problem formulation considers three distinct methods related to maximize the savings function and capacitors to be installed, as well as increasing the voltage profile. This paper uses fuzzy expert system for capacitor location, practical mathematical procedure and particle swarm optimization method for sizing. Initially, the load flow solution for the radial feeder is obtained by Newton Raphson algorithm. Voltage, power factor and real power loss index of distribution system nodes are modeled by fuzzy membership function. Then, a fuzzy inference system containing a set of heuristic rules is designed to determine candidate nodes suitable for capacitor placement in the distribution system. Capacitors are placed on the nodes with highest sensitivity index. The sizing is found by using both practical mathematical procedure and particle swarm optimization. Each method differs from others in terms of fuzzy expert system inputs and sizing methods. The intent of this paper is to study and discuss the results of these methods with reference to savings cost. The proposed methods are tested in IEEE,' IleV, 12 bus & 34 bus test radial feeder system and results simulated from MATLAB.