Energy Management Systems of Microgrids

The distributed operation of parts of the system, denoted as microgrids or, more generally, as local energy communities, could be an effective answer to the issues posed by the increasing complexity of the modern power distribution systems facing the increasing penetration of renewable energy sources and the electrification of urban transportation. The results of the research activities described in the thesis can be divided into three main parts. The first one is the modeling and analysis of low voltage power distribution networks feeding residential, commercial and small-scale industrial consumers including distributed generation units and storage systems. It focuses on an optimization model that has been applied to the energy management system of an experimental microgrid. A mixed integer linear programming model is developed and presented, which takes into account the unbalanced operation of the LV network. The second part focuses on the day-ahead operational planning of a local energy community, which is assumed able to implement transactive energy control actions with allocation of the network power loss. The problem has been addressed by means of two different optimization procedures, namely a centralized mathematical programming model and a specific distributed optimization procedure based on the adoption of the alternating direction method of multipliers (ADMM). The third part is the day-ahead optimization of the operation of a local energy system consisting of photovoltaic units, energy storage systems and loads aimed at minimizing the electricity procurement cost, considering the uncertainties in the load and generation forecasts. Two mixed integer linear programming models are adopted, each for a different representation of the battery: a simple energy balance constraint and the Kinetic Battery Model. The chapter describes the generation of the scenarios, the construction of the scenario tree and the intraday decision-making procedure based on the solution of the multistage stochastic programming.