Reliability Improvement of Distribution System by Optimal Placement of Distributed Generator Using Genetic Algorithm and Neural Network

Abstract

Distributed generator (DG) is now commonly used in distribution system to reduce total power loss and to improve the power quality and reliability of the network. The major task of connecting DG is to identify their optimal placement in the system and to evaluate the amount of power to be generated by DG. By considering this objective, a hybrid technique using Genetic algorithm and Neural-network is proposed in this paper. By placing DG at optimal location and by evaluating generating power based on the load requirement then the number of generators in the network increases and so that different generator states are possible for a particular load condition. The total power loss in the system can be minimized without affecting the voltage stability of the buses. Reliability is old concept and a new discipline. Reliability is, and always has been, one of the major factors in planning, design, operation and maintenance of electric power system. Reliability of an electric supply system has been defined as the probability of providing the user with continuous service of satisfactory quality. Reliability prediction is a method of quantitatively stating what is expected to occur and can be used to indicate the relative merits of alternate design proposals with regard to a predetermined level of performance adequacy. Here considered reliability parameters are Loss of Load probability (LOLP) and Expected Energy Not Supplied (EENS).The LOLP and EENS evaluations are based on peak lo