Announcement
Access to the virtual meeting platform is now closed.
Access to the virtual meeting platform is now closed.
Presentation Details
| Probabilistic forecasts with numerical and machine learning ensembles William B.Hobbs1, Timothy J Silverman2. 1Southern Company, Birmingham, AL, USA.2Photovoltaic Materials and Devices Group, Delft University of Technology, Delft, Netherlands |
Abstract
This work explores the reliability of irradiance forecasts from publicly-available weather prediction model ensembles. That is, verifying how often observed irradiance values occur relative to the probability of them occurring based on output of these ensembles. It will look at conventional Numerical Weather Prediction (NWP) models from NOAA and ECMWF, and will also look at the newest Artifical Intelligence Weather Prediction (AIWP) ensemble model from ECMWF, AIFS ENS, which was specificaly trained for reliability.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.