An Advanced Forecasting System for the Optimum Energy Management of Island Microgrids

Abstract

Variability of energy production is considered to be the main shortcoming in the operation of renewable energy systems. Combination of different Renewable Energy Sources (RES), employment of energy storage and application of Demand Side Management (DSM), are all elements used to encounter the problem of RES variability. Exploitation of such elements in an effective manner challenges the development of advanced Energy Management Systems (EMSs), especially in the case of island microgrids with high shares of RES, lacking the flexibility and capacity of centralized electricity systems to facilitate increased RES penetration. In this work, and in the framework of the Horizon 2020 TILOS project, an advanced Forecasting System (FS) has been developed, able to provide reliable predictions of load demand, wind power and solar power production. The specific variables are independently predicted through a set of forecasting models that produce both deterministic and probabilistic results for different time horizons and time resolutions, fully adjustable to the requirements of any given island microgrid. The developed FS has been deployed and tested considering the smart microgrid of Tilos island, in the SE Aegean Sea, with results obtained demonstrating its ability to provide sufficient and accurate forecasts for all studied variables.