Integration of renewable resources presents a range of very challenging statistical resource characterisation problems, for instance:
- Large scale estimation of multivariate distributions for network planning
- Assessment of renewable resource at times of extreme demand under inevitable limitations of available data
- Estimation of wind forecast error distributions, a non-stationary problem (error distribution depends for instance on prevailing weather system) in which distribution tails are very important
Management of uncertainty in all of these statistical estimation processes is vital, particularly where available data on extremes of demand or forecast errors is very sparse.
This panel will examine the state of the art in this field and discuss key priorities for future research and development, with particular emphasis on management of uncertainty in statistical estimation.