Extreme Learning Machine Based Multi-Agent System for Microgrid Energy Management

Abstract

In this paper, an intelligent energy management system is presented for distributed structure like a smart microgrid. To model the microgrid, a Multi-Agent System is proposed based on Extreme Learning Machine algorithm to estimate the wind and photovoltaic power output from weather data. In this study a microgrid, with different generation units and storage units is considered. Provision of utility grid insertion is also given if the total energy produced by microgrid falls short of supplying the total load or if there is an excess of energy produced instead of to be wasted. Thus the goal of our Multi-Agent System is to control the amount of power delivered or taken from the main grid in order to reduce the electricity bill and make profit by selling the surplus in the energy market. After supplying the load requirements, Extreme Learning Machine algorithm for classification is used to make decision about selling/purchasing electricity from the main grid, and charging/discharging batteries. Finally for simulation, the Java Agent Development Framework platform is used to implement the approach and analyze the results.