Design of photovoltaic and battery systems in the context of Energy Communities through a genetic algorithm approach based on the cost of energy and demand flexibility

In recent years, the growing concern about climate change has significantly heightened, increasing people's awareness of their electricity consumption, and fostering a deeper understanding of the impact of energy production. Furthermore, the increased penetration of Photovoltaic (PV) panels, microturbines, storage systems and the so-called Distributed Energy Resources (DER) indicate a shift in the future role of consumers, that will become prosumer, i.e. simultaneously producing and consuming electricity. As a result of this greater attention, public institutions are advocating a decentralised system in contrast to the traditional one, which consists of a centralised system. On the other hand, the installation of DER encouraged new community-based structures, known as Energy Communities (EC), that are designed to manage the new challenges which are brought by these changes. The goal of this thesis is to analyse the opportunities that are given to these communities and assess their advantages and feasibility. This work presents an optimization model that aims to minimise costs for consumers participating in an EC by designing PV-battery rooftop systems. Taking into account the aforementioned consumer awareness, the thesis investigates also the effect of Demand Management. Firstly, a single-user optimisation is carried out considering a Time-Of-Use (TOU) tariff, in the second part of the analysis the optimisation is subjected to Load Shifting (LS). The model is rooted in a Genetic Algorithm, that determines the sub-optimal size of the PV and the storage as well as its level of charge throughout the week, distinguishing weekdays from the weekends ones. Besides, the work determines the percentages of shiftable load obtained for different time slots that follow the TOU structure. The tool is tested with a dataset consisting of different combinations of time series demand from 50 users, and different PV configurations, considering factors such as different orientations and tilt angles. By identifying the best angles for the PV, coupled with the type of user consumption curve, the aim is to understand the relationship between these features. The effect of load aggregation is also studied by comparing the results of single-user optimisation. Finally, various indexes are evaluated to assess the overall convenience of the system.