Optimal operating scheme of neighborhood energy storage communities to improve power grid performance in smart cities

In Smart Cities (SC), the efficient management of services such as health, transport, public safety, and especially the electricity ensures the welfare of citizens. In recent years, the insertion of renewable sources (RSs) (e.g., solar and wind) in the power grid (PG) of SCs has contributed to meeting the electricity needs of the various consumer units. However, the large-scale integration of these RSs can fatigue the assets, leading to their premature aging and, consequently, compromising the quality of electricity supply. To overcome these challenges, the implementation of Neighboring Energy Storage Communities (NESCs) employing demand response (DR) strategies along with efficient coordination of storage batteries (SBs) could be a promising alternative. In this sense, the present work proposes a mixed-integer linear programming (MILP) model to efficiently manage SBs and the set of household appliances, including charging electric vehicles (EVs), in an NESC provided solely by PG. The proposed model aims to minimize: the total costs related to energy consumption, the peak rebound effect on the total consumption profile, energy wastage through load factor (LF) improvement, and the deep discharges in the SBs during their daily operational cycle. Operational constraints related to the home appliances, such as average usage time, the number of times that the appliance is used daily, etc., are taking into account. The EV state-of-charge (SOC), EV charging rate limits, and initial and final SOC of the SBs, are also considered. A Monte Carlo Algorithm (MCA) is used to simulate the habitual consumption patterns of each customer. The proposed model was implemented in AMPL and solved using CPLEX. The performance of this proposed model is evaluated considering two NESCs differentiated by the number of consumer communities. A first NESC (small-scale) is analyzed considering only two consumer communities. In this NESC, two case studies (Case 1 and 2) are discussed. Next, the second NESC (large-scale) that considers 14 consumer communities is analyzed for the most complete case study (Case 2). Within each NESC, consumer communities are differentiated by the household income and the types of SBs (individual and shared) that support each community. The results corroborate the applicability of the MILP model to real case studies on a diverse scale, guaranteeing the efficient use of PG at the same time that each SB seeks the most optimized operation.