Abstract— The impact of non-scheduled Renewable Energy Sources on distribution networks and the forecasted deployment of Electric Vehicles can lead up to severe power quality stress (i.e. peak demand), in the next future. In order to alleviate these problems, IPP can offer a peak shaving ancillary service to DNOs by means of Energy Storage Systems (ESSs) integration, which represents a technological promising solution. Here, we analyze the sizing of ESSs, installed at the DG connection bus, by Monte Carlo simulations and assess the solutions for a real Italian MV network taking into account several proposed indices related to the ESS cost and size. Index Terms—Ancillary Services, Electric Vehicle, Energy Storage Systems, Renewable Energy Sources, Smart Grid. I. INTRODUCTION The growing of penetration of Distributed Generation (DG) based on non-programmable Renewable Energy Sources (RESs) into Distribution Networks (DN) is determining serious effects in terms of synchronization between power production and energy demand. In particular, distribution systems need to deal with several problems due both to possible reverse power flows on the HV/MV transformer and voltage profiles over the limits imposed by standards. The problems could be exacerbated in the next years up to critical levels for the forecasted deployment of Electric Vehicles (EVs). In fact, as the charging of EVs determines a large consuming of electrical energy, extra-large and undesirable peaks in power demand can deteriorate further the power quality supply in distribution grids. The integration of Energy Storage Systems (ESSs) on DN can be a solution to apply in order to alleviate these operational problems [1], so that the net demand profiles can be equalized, promoting the efficient use of energy. Storage devices can also help to make non-programmable RESs smoother and dispatchable [2], reducing the charges- imbalance as required in some countries. In the context of Smart Grid many technologies and smart control strategies can be implemented on distribution systems by providing ancillary services [3]. In literature several works [4]-[5] investigate the necessity to develop proper control techniques to ensure power delivery to customers and to V. Calderaro, V. Galdi, G. Graber, and F. Lamberti are with the Department of Industrial Engineering, University of Salerno, via Giovanni Paolo II, 132 - 84084 - Fisciano (SA) - Italy (e-mail: vcalderaro@unisa.it, vgaldi@unisa.it, ggraber@unisa.it, flamberti@unisa.it ). G. Graditi is with the ENEA - Research Center of Portici, P. Enrico Fermi, 1 - 80055 Portici (NA) - Italy (e-mail: giorgio.graditi@enea.it ). furnish ancillary services in presence of DG. Of particular interest is the integration of DG with storage systems in order to optimize distribution system operation and maximize the benefits obtainable by the presence of both [6]. Thanks to the ESSs, it is possible to provide ancillary services such as load following, back-up, peak shaving [7], reactive power support [8] and power quality [9]. In this paper, we investigate the sizing of ESS for the introduction of a peak shaving ancillary service that Independent Power Producers (IPPs) offer to Distribution System Operators (DSOs), in order to support the DN during the peak demand period. The goal is achieved by using ESS on board of RES units. ESS can manage the amount of required power to supply customers shaving the load demand during the peak. In this way the DSO can defer infrastructure investments using the power production of RESs when required and paying IPPs for the given service. The energy storage sizing includes the determination of rated power and capacity [10]. The sizing of the ESS affects the costs and the percentage of peak demand that is possible to reduce during the peak. The analysis is performed by Monte Carlo simulations in order to explore the solution space of the problem taking into account several indices for the choice of the ESS. The analysis does not allow finding the optimal solution of the problem but assessing the feasible solutions for providing the peak demand ancillary service. II. METHODOLOGY A. Power System Components The main components of the DG connected distribution network are feeders, loads, DG units and ESSs. Each feeder section is modeled by its series impedance and shunt reactance and the loads by P-Q constant power bus according to a typical hourly load curve. All DG units are considered as photovoltaic (PV) units and, in order to have a daily generation profile, the rated power of each PV unit is multiplied by a typical normalized profile. ESSs are modeled by an ideal voltage generator that supplies/stores constant power at each time step. State of Charge (SoC) is evaluated according to the following equation: I dT C t SoC dT t SoC ESS 3600 1 ) ( ) (    (1) where ESS C [Ah] is the ESS’s capacity, I represents the ESS’s current, which is given by dividing the charge/discharge power of the ESS to its constant voltage, and dT is the time step. Impact assessment of Energy Storage and Electric Vehicles on Smart Grids V. Calderaro, Member, IEEE, V. Galdi, Member, IEEE, G. Graber, Student Member, IEEE, G. Graditi and F. Lamberti, Student Member, IEEE