Multi-objective Strategies for Management and Design of Distributed Electric Storage Systems in a Mediterranean Island M.L. Di Silvestre, M.G. Ippolito, E. Riva Sanseverino, E. Telaretti, G. Zizzo DEIM University of Palermo Palermo, Italy gaetano.zizzo@unipa.it G. Graditi ENEA Portici Research Center ENEA Portici (NA), Italy giorgio.graditi@enea.it Abstract— The paper proposes an analytical study about the design and operation of distributed Electrical Storage Systems(ESS) in an active distribution network supplying a Mediterranean island, according to different criteria. The approach is based on the application of a multi-objective optimization algorithm (NSGA-II) that outputs a set of optimized schedules of the ESS in different configurations when operated according to different criteria (minimum energy losses in the grid, total generation cost, greenhouse gas emissions, limiting the fluctuations of the generation from fossil fuel plants). Moreover the variability of the voltage profile has been investigated also by a comparison with a reference scenario characterized by the absence of storage systems. Keywords— ESS; Multi-objective Optimization; Active Distribution Networks I. INTRODUCTION New active distribution networks are characterized by a deep penetration of Renewable Energy Sources (RES) based generators. In particular, photovoltaic (PV) and winds plants are the most common RES-based generators connected both to low voltage (LV) and medium voltage (MV) distribution networks, also thanks to national support policies (Feed’in Tariffs or Green Tags [1]-[2]) and to the enabling technologies [3]-[4]. Nevertheless, the intrinsic uncertainty of the output power from these generation sources jeopardize their competitiveness in the electricity market and can ultimately limit their commercial success. The variability and the not-dispatchable nature of these sources has led to concerns regarding the reliability and stability of the associated electrical systems, both for grid–connected and for stand–alone systems. In this scenario, Electric Energy Storage Systems (ESS) have a strategic role; in fact a lot of new important applications of these systems have now joined the classical applications like emergency supply for privileged loads or stand-alone systems or supply for small appliances or portable devices. ESSs allow the decoupling in time between electricity generation and utilization, promoting a better penetration of RES-based generation in distribution networks. The ESSs can be used as fast reserve, taking part to the primary regulation in networks where unpredictable loads or generation injections can seriously affect voltage and frequency. According to the most recent Italian technical standards related to the active users connection to distribution grids [5]-[6], RES-based generators are obliged to participate to voltage and frequency regulation, modulating the active and reactive power injections according to specific capability curves. Being able to provide all of these regulation services, ESS are poised to become a fundamental element of the RES-based generators and of the electrical infrastructure of the future [7]-[8]. Moreover, within isolated systems, ESSs have a fundamental importance in maintaining supply continuity in case of lack of production. Finally, suitably programmed charge regulators for ESS allow the optimization of the use of the energy sources implementing load-leveling or peak-shaving functionalities. Other applications for special ESS, integrated inside suitable devices, are aimed at active filtering to improve the Power Quality, protecting loads against perturbations from the grid (voltage sags, harmonic distortions, etc.) or the grid against perturbations from specific disturbing loads. In this context, it is clear that a “smart” design and management of distributed ESS can lead to great advantages for all the distribution networks actors. Considering all these functionalities, the present work proposes a methodological approach to the definition of design and management strategies of distributed ESSs in an automated and isolated active distribution network. The present study analyzes the effect of optimal operation of distributed ESS serving a large distribution network. The test system supplies a Mediterranean island; it is well known that the structure of these systems includes a main diesel generator, the loads and renewable generation sources. In the case under study, it is investigated the possibility to install also some distributed ESS in the network. Such evaluation makes use of a multi-objective optimizer with the following objective functions: ＹＷＸＭＱＭＴＷＹＹＭＰＲＲＳＭＱＯＱＳＯＤＳＱＮＰＰ＠ﾩＲＰＱＳ＠ｉｅｅｅ ＷＶＲＷ