OPTIMAL PLANNING OF DISTRIBUTED GENERATION FOR DISTRIBUTION NETWORK

Abstract

As the power demand in the world is increasing continuously. Therefore there is a need of efficient system along with increased generation. Distributed Generation minimizes the overall cost along with increase in reliability. Distributed Generation improves the voltage profile of the overall system which also leads to the cost reduction. Along with reducing the peak load and cost it is also environment friendly as most of the DG uses renewable energy sources. Therefore to provide reliable and high quality power to customer load there is need of some improvement in older power system modelling. The most common issue now a day is the installation and penetration of DG. DG placement problem is a complex combinational problem having nonlinear objective as well as nonlinear constraints. Therefore there is a need of suitable technique to be applied for the optimal placement of DG. In this thesis work a new approach namely teaching learning combined with harmony search (TLCHS) algorithm for the optimal placement of DG in distribution network for power loss minimization along with voltage profile improvement has been done. In TLCHS algorithm some strategies such as teaching – learning strategy, random mutation, harmony search consideration and local pitch adjustment are employed for maintaining the balance between exploration and exploitation, and the parameters are changed dynamically. The proposed algorithm has been coded in MATLAB. The proposed algorithm has been tested on 33-bus and 69-bus radial distribution network and obtained results has been analysed to see the validity and effectiveness of proposed technique. The effectiveness of this method is more than other older methods .The conventional HS algorithm fails to find the optimal solutions in some cases .to overcome this drawback, the improved versions of HS such as Improved Harmony Search (IHS) and Novel Global Harmony Search (NGHS) have been developed. However, for high dimensional problems, these improved versions of HS has weak searching capacity .In TLCHS algorithm HS is combined with TL as TL based algorithm has improved searching capacity for multi-dimensional problems.