Session: 6.1.2 - Student Competition

Paper Number: 108893

108893 - Evaluating Critical Weather Parameters Using Machine Learning Models 

Wind speed and temperature forecasting accurately for an urban area are two critical elements to mitigate cost and energy in engineering calculations. Recently, researchers use multi-scale numerical assessment models for forecasting weather parameters. They employ coupling a macroscale WRF model with microscale models. In this research, the WRF model results for a 3-day average for parameters such as 10-m wind speed, 2-m air temperature and 2-m relative humidity for two stations in Singapore are reported. The percentage of relative humidity in the morning is at the highest level, while wind speed and the temperature gradually increase in the afternoon and decrease at night. We employ the Recurrent Neural Network (RNN) algorithm to improve the results of the numerical weather model. The results show that the proposed model has a great performance in wind speed forecasting. WRF simulation results agree well with measurement stations, and the RNN algorithm can output more accurate wind speed forecasts. The present research implements a machine learning model to reach highly accurate necessary data in the shortest time. It could pave the way for building engineers to have planning for analyzing wind damage probability and wind hazard mitigation. 

Wind forecasting, WRF, RNN, Numerical weather model 

Wind speed and temperature forecasting accurately for an urban area are two critical elements to mitigate cost and energy in engineering calculations. Recently, researchers use multi-scale numerical assessment models for forecasting weather parameters. They employ coupling a macroscale WRF model with microscale models. In this research, the WRF model results for a 3-day average for parameters such as 10-m wind speed, 2-m air temperature and 2-m relative humidity for two stations in Singapore are reported. The percentage of relative humidity in the morning is at the highest level, while wind speed and the temperature gradually increase in the afternoon and decrease at night. We employ the Recurrent Neural Network (RNN) algorithm to improve the results of the numerical weather model. The results show that the proposed model has a great performance in wind speed forecasting. WRF simulation results agree well with measurement stations, and the RNN algorithm can output more accurate wind speed forecasts. The present research implements a machine learning model to reach highly accurate necessary data in the shortest time. It could pave the way for building engineers to have planning for analyzing wind damage probability and wind hazard mitigation. 

Wind forecasting, WRF, RNN, Numerical weather model 

Wind speed and temperature forecasting accurately for an urban area are two critical elements to mitigate cost and energy in engineering calculations. Recently, researchers use multi-scale numerical assessment models for forecasting weather parameters. They employ coupling a macroscale WRF model with microscale models. In this research, the WRF model results for a 3-day average for parameters such as 10-m wind speed, 2-m air temperature and 2-m relative humidity for two stations in Singapore are reported. The percentage of relative humidity in the morning is at the highest level, while wind speed and the temperature gradually increase in the afternoon and decrease at night. We employ the Recurrent Neural Network (RNN) algorithm to improve the results of the numerical weather model. The results show that the proposed model has a great performance in wind speed forecasting. WRF simulation results agree well with measurement stations, and the RNN algorithm can output more accurate wind speed forecasts. The present research implements a machine learning model to reach highly accurate necessary data in the shortest time. It could pave the way for building engineers to have planning for analyzing wind damage probability and wind hazard mitigation. 

Wind forecasting, WRF, RNN, Numerical weather model 

Presenting Author: Maede Najian Cleveland State University

Presenting Author Biography: Maede is a graduate student in ME program at CSU.