南方湿润地区太阳总辐射估算模型筛选  

Screening with estimated models for global solar radiation in humid regions of South China

作　　者：崔宁博[1,2] 李奇临 李艳灵 吴宗俊 方佩 宋培 李仁笑 CUI Ningbo;LI Qilin;LI Yanling;WU Zongjun;FANG Pei;SONG Pei;LI Renxiao(State Key Laboratory of Hydraulics and Mountain River Engineering, Sichuan University, Chengdu, Sichuan 610065, China;Sichuan Provincial Key Laboratory of Water-saving Agricultural in Hill Area of South China, Chengdu, Sichuan 610066, China)

机构地区：[1]四川大学水力学与山区河流开发保护国家重点实验室,四川成都610065 [2]南方丘区节水农业研究四川省重点实验室,四川成都610066

出　　处：《排灌机械工程学报》2022年第7期693-700,共8页Journal of Drainage and Irrigation Machinery Engineering

基　　金：国家自然科学基金优秀青年科学基金资助项目(51922072);国家自然科学基金资助项目(51779161,51009101);大学生创新创业省级项目(S202110610401,S202110610404,S202110610405)。

摘　　要：为筛选适于模拟中国南方湿润地区太阳总辐射的经验模型,将中国南方地区划分为华东、华中、华南和西南4个区域,选用36个辐射站点逐日数据(1994—2016年),采用最小二乘算法优化3种基于温度模型(T_(1),T_(2)和T_(3)模型)、3种基于日照时数模型(N_(1),N_(2)和N_(3)模型)和2种混合模型(M_(1)和M_(2)模型)的经验参数,并在日、月和年尺度上评价各模型在4个分区的精度.结果表明:在日、月和年尺度上,日照时数模型精度最高,其次为混合模型,温度模型精度最低,日尺度R^(2)分别为0.847~0.898,0.654~0.904和0.435~0.687,月尺度R^(2)分别为0.529~0.857,0.368~0.858和0.334~0.766,年尺度RE为-1.78%~21.86%,-32.88%~24.14%和-37.54%~72.36%;日照时数模型中,N_(2)和N_(3)模型精度最高;混合模型中,M_(2)模型精度最高.因此,N_(2)和N_(3)模型在南方湿润地区模拟效果最好,该研究可为气象资料有限条件下精确计算南方地区太阳总辐射R_(s)提供理论依据.In order to select the optimal experience model for simulating the global solar radiation in humid regions of southern China,Which was divided into four regions,namely,East,Central,South and Southwest China.The daily data of 36 radiation stations(1994—2016)were selected to evaluate the simulation accuracy of three models based on temperature(T_(1),T_(2) and T_(3) model),three models based on sunshine duration(N_(1),N_(2) and N_(3) model)and two hybrid models(M_(1) and M_(2) model)in four regions of southern China.The models were optimized by the empirical parameters with least square method.The accuracy of each model in four regions was evaluated on the daily,monthly and annual scales.The results show that at daily,monthly and annual scales,the accuracy of the sunshine duration models is the highest,followed by the hybrid models and the temperature models.The daily ranges of R^(2) are 0.847-0.898,0.654-0.904 and 0.435-0.687,respectively.The monthly ranges of R^(2) are 0.529-0.857,0.368-0.858 and 0.334-0.766,respectively.The annual ranges of RE are-1.78%to 21.86%,-32.88%to 24.14%and-37.54%to 72.36%,respectively.Among the sunshine duration models,N_(2) and N_(3) models have the highest accuracy.Among the hybrid models,M_(2) model have the highest accuracy.Therefore,the N_(2) and N_(3) models have the best simulation effect in the humid areas of South China.These results can provide a theoretical basis for precisely calculating R_(s) based on limited meteorological data in southern humid region.

关 键 词：太阳总辐射 南方湿润地区 温度模型 日照时数模型 混合模型 

分 类 号：S277.9[农业科学—农业水土工程]