Assessment of Data-Driven Models for Estimating Relative Humidity, case study of Bajgah, Iran
Assessment of Data-Driven Models for Estimating Relative Humidity, case study of Bajgah, Iran
Reza Piraei1 Ali Mohammadi2 Seied Hosein Afzali3
1) PhD Student of Water Recourses Management, Department of Civil and Environmental Engineering, Shiraz University, Shiraz, Iran. E-mail:
2) MSc Student of Water Recourses Management, School of Civil and Environmental Engineering, Tarbiat Modares University, Tehran, Iran. E-mail:
3) Associate Prof. of Civil Engineering, Department of Civil and Environmental Engineering, Shiraz University, Shiraz, Iran. E-mail:
IOI Code
:
XGFB-KHGGC
Publication :
9th.International Congress On Civil Engineering, Architecture & Urban Development(9icsau.ir)
Abstract :
Meteorological variables, particularly relative humidity, play a pivotal role in influencing various aspects of human life and natural environments, spanning agriculture, water resource management, and renewable energy systems. Acknowledging the complexity of developing equations for relative humidity prediction based on meteorological variables, which requires knowledge of the underlying relations between them, this study underscores the advantage of machine learning models that do not rely on explicit equations. Considering limited studies in this domain, this study utilized six machine learning models for predicting daily mean relative humidity, employing historical mean relative humidity, minimum temperature, and maximum temperature from one to three preceding days (resulted from Pearson correlation analysis) as input variables. The models, including Artificial Neural Network, Decision Tree Regressor, Random Forest Regressor, AdaBoost Regressor, Gradient Boosting Regressor, and XGBoost Regressor, are evaluated for their performance using a comprehensive 30-year dataset from Bajgah meteorological station in Iran. The assessment involved the application of six distinct statistical criteria, i.e. RMSE, MAE, MARE, MXARE, R2, and NSE, which highlighted the commendable performance of all models utilized in this study. Based on a ranking method, the Random Forest Regressor, XGBoost Regressor, and Gradient Boosting Regressor models jointly achieved the top rank, with Random Forest Regressor excelling in training data, XGBoost Regressor in test data, and Gradient Boosting Regressor demonstrating robust performance across both datasets. Following these top-performing models, the Artificial Neural Network model secured a subsequent ranking. In contrast, the Decision Tree Regressor and AdaBoost Regressor models exhibited lower performance compared to their alternatives. The study highlights the potential of machine learning models in predicting relative humidity and encourages future exploration of alternative models.
Keywords : hydrology machine learning meteorological variables relative humidity xgboost