Residential Energy Management in Smart Grid: A Markov Decision Process-Based Approach

The deployment of advanced information and communication technologies has helped in the transformation of the traditional power grids into smart grids by introducing demand side management in residential area. The use of demand side management in the residential area can successfully reduce customer´s energy consumption, and can also provide a well balanced energy demand throughout the day. Based on real-time pricing information, a customer can shift his/her energy demand to reduce the energy consumption cost. In this paper, we present a Markov Decision Process (MDP)-based scheduling mechanism for residential energy management (REM) in smart grid. The aim of the proposed work is to reduce the energy expenses of a customer. In this mechanism, the Home Energy Management Unit (HEMU) acts as one of the players, the Central Energy Management Unit (CEMU) acts as another player. The HEMU interacts with the CEMU to fulfill its energy request within its desired budget. The CEMU follows its own dynamic pricing mechanism to decide the price per unit energy for on-peak and off-peak hours. The proposed mechanism is able to reduce the energy expenses of the residential customers.