An alternative learning-based approach for economic dispatch in smart grid system

Abstract

The written report is based on a Final Year Project: An Alternative learning-based approach for Economic Dispatch (ED) in Smart Grid system. The ED problem has existed for many years and many research has been done to improve the approaches to optimize ED. The goal of ED is to minimize the cost of power generation while meeting the load demand and other various constraints. Most traditional Economic Dispatch approaches are computationally expensive due to the fact that most optimization procedures are iterative algorithms. A proposed approach of using Deep Neural Network (DNN) to map the relationship between input power demand and output power generated of Economic Dispatch will be discussed in this report.