Scheduling deferrable appliances and energy resources of a smart home applying multi-time scale stochastic model predictive control

Abstract

In this paper, the problem of scheduling deferrable appliances and energy resources of a smart home (SH) is studied. The SH has a variety of sources that include photovoltaic (PV) panels, diesel generator (DG), and plug-in electric vehicle (PEV) as an energy storage, and also it can transact power with the local distribution company (DISCO). Herein, the appliances of the SH are categorized into non-deferrable and deferrable appliances, and also the deferrable appliances are divided into two groups with hour-scale and day-scale deferrable features. The challenges of the problem include modeling the economic and technical constraints of sources and appliances and addressing the variability and uncertainties concerned with the power of PV panels that make the problem a mixed-integer nonlinear programming (MINLP), time-varying (dynamic), and stochastic optimization problem. In this study, a multi-time scale stochastic model predictive control (MPC) and a combination of genetic algorithm (GA) and linear programming (GA-LP) are applied to address the above mentioned issues and solve the problem. The numerical study demonstrates the competence of multi-time scale approach in the stochastic MPC, and also the proficiency of proposed approach for hour-scale and day-scale deferrable appliances scheduling.