Modeling forecast uncertainty using fuzzy clustering

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F13%3A86096970" target="_blank" >RIV/61989100:27240/13:86096970 - isvavai.cz</a>
Result on the web
<a href="http://dx.doi.org/10.1007/978-3-642-32922-7_30" target="_blank" >http://dx.doi.org/10.1007/978-3-642-32922-7_30</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-642-32922-7_30" target="_blank" >10.1007/978-3-642-32922-7_30</a>

Result language
angličtina
Original language name
Modeling forecast uncertainty using fuzzy clustering
Original language description
Numerical Weather Prediction (NWP) systems are state-of-the-art atmospheric models that can provide forecasts of various weather attributes. These forecasts are used in many applications as critical inputs for planning and decision making. However, NWP systems cannot supply any information about the uncertainty of the forecasts as their immediate outputs. In this paper, we investigate the application of Fuzzy C-means clustering as a powerful soft computing technique to discover classes of weather situations that follow similar forecast uncertainty patterns. These patterns are then utilized by distribution fitting methods to obtain Prediction Intervals (PIs) that can express the expected accuracy of the NWP system outputs. Three years of weather forecast records were used in a set of experiments to empirically evaluate the applicability of the proposed approach and the accuracy of the computed PIs. Results confirm that the PIs generated by the proposed post-processing procedure have a h
Czech name
—
Czech description
—

Publication year
2013
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)