The forecasting system is innovative because it generates deterministic and uncertainty forecasts for the next 60 hours, with a time resolution between 15 and 60 minutes. The system also combines, in real time, spatially distributed measurements collected by smart devices (such as meters) and sensors, in order to model the effect that solar generation has on clouds. These innovative methods have recently been addressed in papers published in international journals [1,2]. This model is also capable of dealing with a high and continuous flow of data, and contains a module to extrapolate solar generation to produce forecasts at the transformer substation. Lastly, it contains modules to manage warnings in the event of communication failures, quality control and pre-processing of data.

The forecasting system is now operational and generating forecasts, for each hour, for 29 transformer substations and 30 microgenerators in Évora, Portugal.

The CPES team involved in the demonstration phase is composed of Ricardo Bessa, Jorge Filipe and Wagner Franchin.             

Publications

[1] R.J. Bessa, A. Trindade, V. Miranda, “Spatial–temporal solar power forecasting for Smart Grids,” IEEE Transactions on Industrial Informatics, vol. 11, no. 1, pp. 232-241, Feb. 2015. [open access]

[2] R.J. Bessa, A. Trindade, C.S. Silva and V. Miranda, “Probabilistic solar power forecasting in smart grids using distributed information,” International Journal of Electrical Power & Energy Systems, vol. 72, pp. 16-23, Nov. 2015.

The INESC TEC researchers mentioned in this article are associated with the following partner institutions: UP-FEUP and INESC TEC.

INESC TEC, July 2015