首页|Time-Series Embeddings from Language Models:A Tool for Wind Direction Nowcasting

Time-Series Embeddings from Language Models:A Tool for Wind Direction Nowcasting

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
Wind direction nowcasting is crucial in various sectors,particularly for ensuring aviation operations and safety.In this context,the TELMo(Time-series Embeddings from Language Models)model,a sophisticated deep learning ar-chitecture,has been introduced in this work for enhanced wind-direction nowcasting.Developed by using three years of data from multiple stations in the complex terrain of an international airport,TELMo incorporates the horizontal u(east-west)and v(north-south)wind components to significantly reduce forecasting errors.On a day with high wind direction variability,TELMo achieved mean absolute error values of 5.66 for 2-min,10.59 for 10-min,and 14.79 for 20-min forecasts,processed within a swift 9-ms/step timeframe.Standard degree-based analysis,in comparison,yiel-ded lower performance,emphasizing the effectiveness of the u and v components.In contrast,a Vanilla neural net-work,representing a shallow-learning approach,underperformed in all analyses,highlighting the superiority of deep learning methodologies in wind direction nowcasting.TELMo is an efficient model,capable of accurately forecast-ing wind direction for air traffic operations,with an error less than 20° in 97.49％of the predictions,aligning with re-commended international thresholds.This model design enables its applicability across various geographical loca-tions,making it a versatile tool in global aviation meteorology.

wind nowcastingwind componentswind directiontime series predictiondeep learning

Décio ALVES、Fábio MENDON?A、Sheikh Shanawaz MOSTAFA、Fernando MORGADO-DIAS

展开 >

University of Madeira,Campus Universitário da Penteada 9020-105 Funchal,Portugal

Interactive Technologies Institute/Laboratory for Robotics and Engineering Systems and Agência Regional para o Desenvolvimento da Investigação,Tecnologia e Inovação,Edif.Madeira Tecnopolo,Caminho da Penteada piso-2,9020-105 Funchal,Portugal

Interactive Technologies Institute/Larsys/Funda??o para a Ciência e a TecnologiaInteractive Technologies Institute/Larsys/Funda??o para a Ciência e a TecnologiaInteractive Technologies Institute/Larsys/Funda??o para a Ciência e a TecnologiaAgência Regional para o Desenvolvimento da Investiga??o,Tecnologia e Inova??o,and Portuguese Technical Engineering Order(OET)

10.54499/LA/P/0083/202010.54499/UIDP/50009/202010.54499/UIDB/50009/2020

2024

10.1007/s13351-024-3151-9

气象学报(英文版)
中国气象学会

气象学报(英文版)

CSTPCD
影响因子：0.57
ISSN：0894-0525
年,卷(期)：2024.38(3)