A Comparative Study on Genetic Algorithm and Mixed Integer Linear Programming based Optimal Home Energy Management System

This paper presents a Home Energy Management System (HEMS) model consisting of schedulable and non-schedulable loads supplied from the grid and solar PV system under a dynamic tariff structure. The objective is to minimize the electricity consumption cost and maximize the independence from the grid. Mostly, the two objectives are conflicting, and hence the Weighted Sum Method (WSM) based Multi-Criteria Decision Making (MCDM) is employed to develop a single objective function in order to achieve a compromised solution. Mixed Integer Linear Programming (MILP) and Genetic Algorithm (GA) techniques are implemented on the HEMS to optimize the mentioned objective. A detailed comparative analysis of these two optimization techniques is presented for single and multi-objective optimization problems. The analysis elucidates that the MILP performs better in terms of increased cost minimization of about 15 %, reduced average grid power, and increased Independence Performance Index (IPI) of 0.3546 when compared to the unscheduled case.