FOCUSED-Short-Term Wind Speed Forecast Correction Algorithm Based on Successive NWP Forecasts for Use in Traffic...

Area of Science:

• Meteorology and atmospheric science
• Artificial intelligence and machine learning
• Transportation engineering

Background:

• Short-term wind speed prediction is crucial for traffic management, particularly in coastal areas prone to hazardous bora gusts.
• Current decision support systems exhibit occasional ambiguities in forecasting high-speed wind conditions, impacting traffic safety.
• Accurate wind forecasting is essential for mitigating risks associated with extreme weather events in transportation.

Purpose of the Study:

• To introduce and evaluate a novel algorithm, FOCUSED (FOrecast Correction Using Successive prEDictions), for enhancing short-term wind speed forecasts.
• To improve the prediction accuracy of bora gusts, which pose significant risks to road traffic.
• To provide a more reliable tool for traffic management systems operating in challenging coastal wind conditions.

Main Methods:

• The FOCUSED algorithm iteratively forecasts wind speed multiple times for the same future period using numerical weather prediction model characteristics.
• Iterative predictions serve as input features for the FOCUSED algorithm to generate corrected wind speed forecasts.
• Performance evaluation involved comparing FOCUSED against artificial neural networks, random forests, support vector machines, and linear regression.

Main Results:

• The FOCUSED algorithm demonstrated superior performance in wind speed forecast correction compared to traditional machine learning methods.
• The study utilized a comprehensive dataset including five years of real-world wind measurements from the Krk bridge, Croatia.
• Complementary historical forecasts from the ALADIN (Aire Limitée Adaptation dynamique Développement InterNational) numerical weather prediction model were incorporated.

Conclusions:

• FOCUSED offers a significant improvement in short-term wind speed forecasting accuracy, particularly for extreme events like bora gusts.
• The algorithm's iterative prediction approach effectively corrects forecast errors, enhancing its practical applicability.
• The findings suggest FOCUSED can substantially improve the reliability of traffic management systems in hazardous wind conditions.