26–27 Feb 2020
RMIB
Europe/Brussels timezone

Statistical post-processing of wind speed forecasts using convolutional neural networks

Not scheduled
20m
Conference Room (RMIB)

Conference Room

RMIB

Ringlaan 3 B-1180 Brussels Belgium

Speakers

Mr Simon Veldkamp (KNMI; Utrecht University)Dr Maurice Schmeits (KNMI)

Description

Weather forecasts as provided by numerical weather prediction (NWP) models such as Harmonie-Arome, which is used by KNMI, are often (too) deterministic. Because of uncertainty in the forecast, it is preferable to have a full probability distribution instead.
Current statistical post-processing methods for providing a probabilistic forecast are not capable of using full spatial patterns from the NWP model. Recent developments in deep learning (notably convolutional neural networks) have made it possible to use large gridded input data sets. This could potentially be useful in statistical post-processing, since it allows us to use more spatial information. In this study we consider wind speed forecasts for 48 hours ahead, as provided by the Harmonie-Arome model. Convolutional neural networks, fully connected neural networks and Quantile Regression Forests are used to obtain probabilistic wind speed forecasts. Comparing these methods shows that Convolutional neural networks improve on the other methods, especially for medium to higher wind speeds.

Primary authors

Mr Simon Veldkamp (KNMI; Utrecht University) Dr Maurice Schmeits (KNMI) Dr Kirien Whan (KNMI)

Presentation materials

There are no materials yet.