489 978-1-7281-0339-6/19/$31.00 ©2019 IEEE An Optimal Day-Ahead Operation Strategy for Hybrid Energy Microgrid Mohamed Elgamal 1 , Nikolay V. Korovkin 2 , Ahmed Refaat 3 Institute of Energy and Transport Systems Peter the Great Saint-Petersburg Polytechnic University Saint-Petersburg, Russian Federation 1 elgamal.mm@edu.spbstu.ru, 2 Nikolay.Korovkin@gmail.com, 3 Ahmed_refaat_1984@eng.psu.edu.eg Akram Elmitwally Electrical Engineering Department Najran University Najran, Saudi Arabia kelmitwally@yahoo.co Abstract— This paper proposes a strategy for optimal day- ahead operation of a microgrid. The latter includes hybrid energy resources and an energy storage system (ESS). The forecasted day-ahead hourly average of metrological data and loads are fed into the energy management system (EMS). Accordingly, it decides the day-ahead hourly active and reactive power shares of each energy source. It also identifies the ESS charging/discharging periods and the tap setting of the main grid coupling transformer. The overall objective is to maximize the microgrid profit satisfying all constraints. The microgrid purchases/sells active and reactive powers from/to the main grid with time-varying energy price. The day-ahead operation of the microgrid is formulated as an optimization problem solved by a combined rule base - heuristic approach. The modified particle swarm optimization (PSO) technique is used as optimization solver. Moreover, the efficacy of the proposed EMS is verified by performance comparison to recent literature. Keywords— Energy management; Optimization; Renewable generation; Energy storage I. INTRODUCTION Microgrid (MG) is a distribution grid integrating different types of distributed generators (DGs) to supply loads. It can work in a grid-connected or islanded configurations[1]–[3]. Various ESSs and DGs in the MG such as photovoltaic units (PV), wind turbine units (WT), fuel cell units (FC), and microturbine units (MT) have to be operated in a coordinated manner. So, an EMS has to be established [4]. EMS supervises the dynamic operation of electrical generation and transmission systems to maintain security of energy supplies at the minimum cost [5]. It performs many functions such as monitoring, processing, and communicating/negotiating. EMS may also estimate generated powers, load levels, and energy prices of market. Thus, EMS optimizes MG performance in view of the technical and economic constraints [4]. The control framework of MG EMS is categorized into centralized, hierarchical and decentralized EMSs [5]. In centralized EMS, the central controller collects all the system data such as generating power of DGs, operating cost, and demand. Then, it determines the optimal control decisions for all MG subsystems. In hierarchical EMS, the main MG controller exchanges the information with each local controller in real-time. The MG central controller figures out the optimal decision and informs the local controllers. The main advantage of centralized and hierarchical types is that they may get the global optimal solution. Decentralized EMS is formed by distributed controllers with coordinated operation. The common approaches used in centralized EMS include PSO, meta-heuristics algorithms such as BAT algorithm, hyper-spherical search algorithm, mathematical programming, Petri-net method, and artificial intelligence methods [6]. Refs. [7]–[21] report EMSs for MG based on various optimization techniques to minimize the operating cost and load shedding. These include only the active power dispatch of the energy resources. So, this may not yield the maximum possible profit of the MG. Also, it may bring voltage violations and line congestion problems. Ref. [6] presents a PSO-based EMS that considers both active and reactive power dispatch of the DGs with the main grid. Nonetheless, it ignored the use of devices such as under load tap changer of the main grid transformer (ULTC) and available capacitor banks to eliminate voltage violations and line congestions. This paper proposes a day-ahead EMS for a grid-connected MG using a combined rule base-PSO method. The proposed EMS has the following merits: • Considers the reactive power cost and dispatch of DGs in the EMS, • Takes the reactive power exchange between the MG and the main grid into account, • Maintains buses voltages, eliminates lines congestions, and prevents overstress on the ULTC. II. PROBLEM FORMULATION In this paper, DGs and ESS are optimally operated to maximize the day-ahead profit of MG. Both the active and reactive powers are involved. The profit of MG (MGPRO) is expressed as: