Affiliation:
1. MoES: Ministry of Earth Sciences
2. Maulana Abul Kalam Azad University of Technology West Bengal
Abstract
Abstract
In this study, we have attempted to determine the wind characteristics of Netaji Subhas Chandra Bose International Airport (NSCBIA) and predict wind speed and direction 9, 30 hours ahead using a new machine learning (ML) technique. For this, we have collected METAR wind speed and direction data from the NSCBIA for the period 2016-2021. On analysis of the wind speed and direction data, it is observed that a substantial amount of calm wind exists in NSCBIA. The wind speed varies throughout the day with a maximum around 09-11 UTC. The maximum randomness in wind direction is observed around the 06-09 UTC. The maximum wind speed is observed during April-July. Minimum wind speed is observed in December and January. Generally, wind speed remains low during 23-00 UTC but for winter months (October-January) low wind speed is observed during 15-21 UTC. MAE, RMSE and SSE are low for SSA-LSTM in predicting the wind speed. MAE, RMSE, and SSE are higher in predicting wind direction but within tolerance limits owing to wind direction randomness.
Publisher
Research Square Platform LLC