Day ahead microgrid optimization: A comparison among different models

The paper deals with a LV grid-connected microgrid located in the Savona Campus of Genoa University and called Smart Polygeneration Microgrid (SPM). Such microgrid is designed to produce electric and thermal energy by means of renewable generation units, co-generative gas turbines, a storage unit and electric vehicle charging stations. The work fo-cuses its attention on the issue of the microgrid day-ahead power production optimization. Such problem consists in finding the power production profiles for all the dispatchable units in order to optimize a well-defined goal (economical, environmental, etc). This requires to model all the components of the grid in order to define the set of constraints for the optimiza-tion procedure. In particular, the way of representing the electric power balance equations plays a crucial role since a trade-off between accuracy and computational efforts is required. Different approaches are presented and tested on the SPM, e.g. complete load flow, modified DC load flow and single bus-bar representation, in order to highlight the peculiarity of the different optimization models obtained and evaluate their impact on computational time, re-sult accuracy and possibility of future experimental validation.