1.
Algorithm for short-time correction of wind speed forecasting modelsZdravko Kunić, 2023
Opis: In this dissertation, a new algorithm for forecast correction of short-term wind speed predictions is proposed to improve the forecast of bora gusts, frequently resulting in high-speed wind situations dangerous for traffic. The motivation arises from occasionally ambiguous results of the decision support system relying only on the last forecast, which aids traffic management in strong and gusty wind conditions on the roads in Croatia. The proposed correction algorithm uses numerical weather prediction model characteristics to iteratively forecast the wind speed multiple times for the same future window. Iterative predictions are used as input features of the algorithm, and corrected predictions result from the output. The proposed algorithm is tested with artificial neural networks, random forests, support vector machines, and linear regression to demonstrate the superiority of the algorithm performance on a dataset comprising five years of actual data measurements at the Croatian bridge Krk and complementary historical forecasts by ALADIN numerical weather prediction model.
Najdeno v: osebi
Ključne besede: upravljanje prometa, napoved hitrosti vetra, popravek napovedi, nevronske mreže, zaporedne napovedi
Objavljeno: 09.11.2023; Ogledov: 339; Prenosov: 25
Celotno besedilo (3,75 MB)
