Estimating Daily Dew Point Temperature Based on Local and Cross-StationMeteorological Data Using CatBoost Algorithm  

作　　者：Fuqi Yao Jinwei Sun Jianhua Dong 

机构地区：[1]School of Hydraulic Engineering,Ludong University,Yantai,264010,China [2]State Key Laboratory ofWater Resources and Hydropower Engineering Science,Wuhan University,Wuhan,430072,China

出　　处：《Computer Modeling in Engineering & Sciences》2022年第2期671-700,共30页工程与科学中的计算机建模（英文）

基　　金：the Shandong Provincial Natural Science Fund(ZR2020ME254 and ZR2020QD061).

摘　　要：Accurate estimation of dew point temperature(Tdew)plays a very important role in the fields of water resource management,agricultural engineering,climatology and energy utilization.However,there are few studies on the applicability of local Tdew algorithms at regional scales.This study evaluated the performance of a new machine learning algorithm,i.e.,gradient boosting on decision trees with categorical features support(Cat Boost)to estimate daily Tdew using limited local and cross-station meteorological data.The random forests(RF)algorithm was also assessed for comparison.Daily meteorological data from 2016 to 2019,including maximum,minimum and average temperature(Tmax,Tmin and Tmean),maximum,minimum and average relative humidity(RHmax,RHmin and RHmean),maximum,minimum and average global solar radiation(Rsmax,Rsmin and Rsmean)from three weather stations in Hunan of China were used to evaluate the CatBoost and RF algorithms.The results showed that both algorithms achieved satisfactory estimation accuracy at the target stations(on average RMSE=1.020℃,R^(2)=0.969,MAE=0.718℃and NRMSE=0.087)in the absence of complete meteorological parameters(with only temperature data as input).The Cat Boost algorithm(on average RMSE=1.900℃and R^(2)=0.835)was better than the RF algorithm(on average RMSE=2.214℃andR^(2)=0.828).The accuracy and stability of the CatBoost and RF algorithms were positively correlated with the number of input parameters,and the three-parameter algorithms achieved higher estimation accuracy than the two-parameter algorithms.The developed methodology is helpful to predict Tdew at regional scale.

关 键 词：Dew point temperature categorical boosting random forests cross-station accuracy 

分 类 号：TN9[电子电信—信息与通信工程]