Techno-Economic Framework for Optimal Capacity Expansion of Active Microgrid in the Mediterranean: A Case Study of MCAST

Abstract

The purpose of microgrids is to improve system flexibility and resilience during normal and emergency conditions. The ceaseless load growth mandates to increase microgrid's capacity, thereby improving the system flexibility and resilience. However, capacity expansion requires significant investments, making it essential to identify the optimal capacity of energy resources. The methodologies proposed in the literature identifies the microgrid's capacity with an assumption of investments with a single installment. This way of theoretical approach leads to unrealistic solutions. Besides, microgrid's participation in a flexible market will enhance its performance both in commercial and technical aspects. Therefore, this paper proposes a realistic framework with the concept "expansion through time" inspired by "Real Options Theory." This framework includes practical parameters like resource & load uncertainty, physical space required to install, revenue generated by resources, and maximum demand penalty, on top of electrical parameters; constrained with significant return in investments to improve the overall savings. In addition, this paper proposes a market participation model for microgrid, which defines a bidding process with two components, such as regular and flexible portions under both normal and extreme conditions. This study considers renewable-based energy resources like solar-photovoltaic plants (SPPs) and battery energy storage systems (BESSs) as microgrids' energy resources. The system chosen for testing the efficacy of the proposed framework is a real-world active-microgrid of Malta College of Arts, Science and Technology (MCAST), located on an island.

Economic and Regulatory Uncertainty in Renewable Energy System Design: A Review

“…For instance, optimization models have been embedded in ROA [24] for auxiliary purposes (determining the optimal operation of a battery [24] or calculating the prospects of smart grid diffusion [36,39,127]). Or optimization models are reformulated to include flexibility, simulating "ROA thinking" [86,119]. Only one genuinely hybrid approach [66] effectively combines ROA and optimization to simultaneously calculate the optimal sizing and investment timing of a battery system.…”

“…If it is not possible or not desirable to linearize the problem, a more complex solver, which often leads to sub-optimal solutions, must be used. Metaheuristic algorithms such as genetic [21,43], particle swarm [55,89], grey wolf [119], and non-linear solvers such as deterministic policy gradient [66], DICOPT [138], ANTIGONE [17], and KNITRO [107] have been employed in the literature.…”

Design and Performance Investigation of a Pilot Micro-Grid in the Mediterranean: MCAST Case Study

“…The ultimate aim is for the micro-grid to isolate from the main grid even if it means that the controller has to shed non-critical loads to save Currently, the DERs of the electrical system are made up of 63.24 kWP of PV systems. These are distributed on the three buildings, having 21.12 kWP each on both Blocks J and D and 21 kWP on Block F [12]. We propose enhancing the reliability of the micro-grid with a battery energy storage system (BESS) so that the micro-grid system will be able to isolate for a period of time partially.…”

“…These penalties vary for different countries and organizations. According to EC, there is a penalty for every ton of CO 2 emitted above a given threshold [10][11][12][13][14][15][16][17][18][19][20][21][22][23].…”

Probabilistic Techno-Economic Design of Isolated Microgrid

“…The techno-economic design framework with the market participation model for the optimal capacity expansion of PV/Battery configuration in a grid connected MG is discussed in [31]. The randomness of the 24-hourly annual average solar irradiation variation from a single year is represented using the beta distribution function and the variation of minute wise 24h annul average load demand is represented using a normal distribution.…”