On-line Economic Dispatch of Distributed Generation Using Artificial Neural Networks

Abstract

In recent years, distributed generators (DG) are most widely installed in distribution system to meet the increasing demand and especially to reduce the losses. According to demand, dispatch of generator should be modified for economic operation. The Economic Dispatch (ED) of DGs are usually solved by conventional methods such as Lambda iteration method, Dynamic Programming etc., or any optimization technique such as Genetic algorithm (GA), Evolutionary Programming (EP) etc., This off-line methods of solving ED problem require comparatively large computation time and are not suitable for on-line applications. Therefore, it is important to estimate Real Power dispatch values within a short period. This paper presents an On-line ED of various non-renewable DGs for various demands using Artificial Neural Networks namely Back Propagation Neural Network (BPNN) and Radial Basis Function Neural Network (RBFNN). The input pattern for Neural Networks (NN) is demand and output is corresponding optimal real power dispatch. The input and output patterns for NN is obtained using evolutionary programming method. In this work two diesel engines and two fuel cells are used as DG. This case study has been illustrated in a distribution system having two types of four numbers of DGs. The test result shows that the proposed method is better for real time ED.