Abstract
In recent years there have been growing concerns over pollution, energy security, rising energy prices and in particular how greenhouse gas emissions are contributing to climate change—the world’s reliance on fossil fuels for generating electricity is unsustainable. Renewable energy is now broadly recognized as the critical path towards future sustainability and reducing the environmental footprint. Renewable energy sources such as wind and solar are inherently intermittent, and so their integration into an electric power grid requires accurate and reliable estimation of uncertainties. This research present a new data-driven method for constructing a short-term probabilistic forecast of solar irradiation, based on a nonparametric bootstrap technique and map of sun positions accounting for the systematic heteroscedasticity of solar irradiation.
The new data-driven method delivers sharp and calibrated ensembles of one-hour forecasts and can be altered to cater for any predictive horizon from five minutes to a few hours. Our probabilistic forecast will allow one to both assess a wide range of uncertainties in solar energy and facilitate decision-making.
Click here to see all available video seminars.
Click here to go to the SPREE HOMEPAGE.
Brief Bio
Adrian has recently completed his PhD in Applied Mathematics from the University of South Australia. His thesis title was “Mathematical tools for maximizing renewable energy use in electricity supply”. Adrian’s research interests include: time series analysis, probabilistic forecasting, battery storage, and anything else related to the utilisation of renewable energy.
