基于云辐射强迫的太阳辐射短期预报模型  

A Short-Term Solar Irradiance Forecast Model Based on the Relative Cloud Radiative Forcing

作　　者：姚德贵 刘唯佳 韩永翔[3] 李哲 梁允 YAO Degui;LIU Weijia;HAN Yongxiang;LI Zhe;LIANG Yun(State Grid Henan Electric Power Research Institute,Zhengzhou 450000,Henan,China;Hangzhou Meteorological Bureau,Hangzhou 310051,Zhejiang,China;Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters,Nanjing University of Information Science and Technology,Nanjing 210044,Jiangsu,China)

机构地区：[1]国网河南省电力公司电力科学研究院,河南郑州450000 [2]杭州市气象局,浙江杭州310051 [3]南京信息工程大学气象灾害预报预警与评估协同创新中心,江苏南京210044

出　　处：《电网与清洁能源》2023年第1期85-90,99,共7页Power System and Clean Energy

基　　金：国家自然基金面上项目(41875176)。

摘　　要：太阳能持续性模型常被用作基准模型以评估其他模型的预报性能,但有云时的预报准确率很低。基于云相对辐射强迫(RCRF)的持续性,构建了RCRF模型。它能更好地预报有云时段的太阳辐照度。利用美国南部大平原中心站长达16年的太阳辐射资料,通过一个有云个例和所有样本评估了新模型的预报性能,并与应用最广泛的Simple模型作对比。结果表明:新模型的预报性能优于Simple模型,百分比误差评分(S)评估结果显示新模型6 h预报精度相较于Simple模型最大可提高至56%,预报时效也延长了0.25~2 h不等。RCRF模型为太阳能预报模式提供了准确率更高的基准模型。Persistence models are often used as benchmark models to evaluate the forecast performance of other models in predicting solar irradiances,but the forecast accuracy remains less accurate under cloudy conditions.Based on the persistence of relative cloud radiative forcing(RCRF),an RCRF model is constructed to better forecast solar irradiances during cloudy periods.The model performance is evaluated with a cloudy individual case and all samples observed,and further compared with the most widely used Simple persistence model by the use of the solar irradiance data up to 16 years from the Southern Great Plains at the Central Facility site in the U S.The results show that the new model outperforms the Simple model,and the6 h forecast accuracy of the new model can be improved by up to56%compared to the Simple model based on the percent error score(S),and the corresponding forecast lead time is extended to 0.25 h to 2 h.The RCRF model provides a higher benchmark for solar forecasting models.

关 键 词：持续性模型 太阳辐照度 云相对辐射强迫 预报准确率 

分 类 号：TK511[动力工程及工程热物理—热能工程]