Toward an intelligent community microgrid energy management system based on optimal control schemes

Abstract

This article investigates a modelization of the energy management strategy of a grid-tied photovoltaic (PV), wind generator, and battery system. The model uses a smart metering system to collect data from various electrical components of a microgrid. The power optimization strategy is modeled under demand response, including time-of-use pricing to create a home's energy flow more efficiently. Two optimal control strategies are developed to coordinate energy expenses, expressed as the cost of minimizing electricity supplied by the utility grid and maximizing the opportunity for energy to sell to the grid. A nonlinear open-loop control and quadratic programming using model predictive control (MPC) are compared to assess the performance of the developed model. The design model takes into consideration the operational cost of batteries and degradation. The importance of the energy storage system is demonstrated in the context of energy conservation and gain on the demand side. MPC's capacity to specify the proper control and forecast future system reaction enables the proposed method to ensure significant daily energy and cost savings from 70.44 to 24.74 $ day−1 in the first scenario (without disturbances) and from 70.44 to 18.24 $ day−1 in the second case (with disturbances). In both methods, the net savings of greenhouse gas are approximately 288.10 (kgCO2-eq) and 338.87 (kgCO2-eq). These show better performance of the MPC compared to the linear model. Highlights Investigate the dynamic performance of an intelligent home energy management scheme. Implement a grid-tied photovoltaic, wind turbine, and energy storage system. Use intelligent metering to collect data from various electrical components of a smart home. Compare the energy saving from an open-loop algorithm and a model predictive control. Assess the topmost benefits of the optimal control method of grid-tied distributed energy resources.