A mixed-integer linear programming approach for the optimization of residential PV-battery energy storage system

Europe´s electricity sector is facing a major historical turning point shifting away from fossil fuels towards more sustainable energy sources and moving from vertically integrated public monopolies into competitive private companies in unbundled and liberalized markets. In this scenario consumers are expected to play a fundamental role in realising the full potential of the European energy market; the EU energy strategy, with new policy and regulatory initiatives, in fact, recognizes consumers and communities as a key driver of this process, encouraging them to take full ownership of the energy transition and empowering them to actively participate in the electricity market by generating, consuming and selling electricity back to the grid and interacting with other energy market participants. Citizens are in fact no longer restricted to the role of passive end-use consumers, but they are asked to be energy producers, or ‘prosumers’, representing an important contribution to global sustainability and helping to decarbonize the electricity sector. Furthermore, as well as allowing to increase the amount of renewable energy generation, they have the potential to reduce the energy supply-demand gap and electricity system losses and to provide opportunities for demand-side management and for an active grid support increasing grid reliability flexibility, and resiliency. To this end, distributed energy storage systems at the residential level have been identified as a priority technology to open up new possibilities for local flexibility solutions and participation in demand response, leading the energy transition towards new energy configurations, such as self-generation and self-consumption schemes and peer-to-peer selling of the self produced energy. Combining solar power generators and battery storage is one of the most common ways of reaching self-sufficiency in residential buildings by increasing the grid independence of individual households. Due to their cost and growth perspective, battery storage coupled to photovoltaic (PV) generation systems have reached a good level of competitiveness and market penetration in many European countries and they will increase as more such systems become available. In this thesis a battery energy storage systems (BESS) coupled with grid-connected rooftop-mounted residential photovoltaic generation is analyzed. The multi apartment building is located in Civitavecchia (Rome), central Italy, with ten households living in rent at a subsidized price and it is covered by the energy retrofit intervention and building renovation plan set by “Civitavecchia SMART-A.T.E.R.” project aiming to convert the old buildings in new Near Zero Energy Buildings (NZEBs) and to create energy communities. The aim of this thesis is to optimize the sizing and the operation of the battery energy storage system so as to maximize the households’ self-consumption and minimize their electricity bill, whilst ensuring the correct charge-discharge cycles scheduling strategy in order to assure better performance and a longer lifetime of the batteries. To this end, a multi-objective mixed-integer linear programming (MILP) formulation is proposed and it is solved by CPLEX. The battery sizing and operational parameters, in terms of number of batteries and charging/discharging operation mode, are included in the optimization problem and a penalty coefficient is imposed to limit the number of batteries needed. Collected and estimated data of potential photovoltaic production and households’ demand profiles are used to optimize the battery storage system using hourly dataset for different seasons