Peder Bacher

TL;DR: In this paper, a two-stage method is proposed to forecast hourly values of solar power for horizons of up to 36 h. The results indicate that for forecasts up to 2 hours ahead, the most important input is the available observations of PV power, while for longer horizons numerical weather predictions (NWPs) are the more important input.

TL;DR: The method is suited to online forecasting in many applications and in this paper it is used to predict hourly values of solar power for horizons of up to 36 h, where the results indicate that for forecasts up to 2 h ahead the most important input is the available observations ofSolar power, while for longer horizons NWPs are theMost important input.

TL;DR: In this paper, a hierarchy of models of increasing complexity is formulated based on prior physical knowledge and a forward selection strategy is suggested enabling the modeller to iteratively select suitable models with increasing complexity, with which building characteristics such as thermal conductivity, heat capacity of different parts, and window area, are estimated.

TL;DR: A non-parametric machine learning approach used for multi-site prediction of solar power generation on a forecast horizon of one to six hours and shows competitive results in terms of root mean squared error on all forecast horizons.

TL;DR: In this article, the authors presented a method for forecasting the load for space heating in a single-family house using data from sixteen houses located in Sonderborg, Denmark, combined with local climate measurements and weather forecasts.