摘要
风电在我国能源结构转型中具有重要地位,但其波动性也带来严峻挑战。数值模式预报的风速数据是风电出力预测和高效消纳的重要基础,因此需要评估不同模式的预报效果。本文通过对比分析4种主流数值模式的风速预报效果,全面评估它们在我国冬季不同区域和不同条件下的预报精度,以期为我国冬季大风期风速预报提供参考。基于不同分辨率、不同初始场、不同同化方案的4种数值预报模式,结合我国131个站点观测资料,本文对预报风速的误差分布特征与预报能力进行了研究与分析;同时聚焦典型站点,分析了不同风速段、不同区域的预报误差特征及预报能力。研究结果表明:集合预报模式的预报结果在复杂地形条件下更科学;高分率单一模式对简单下垫面的风速波动性预报较好;白天预报效果好于夜间;平原风速预报效果最好。
Wind power plays a crucial role in China's energy structure transition,yet its variability poses significant challenges.Numerical model forecasted wind speed data serves as a vital foundation for wind power output prediction and efficient integration.Therefore,it is imperative to evaluate the forecasting performance of different models.This study comprehensively assesses the wind speed forecasting accuracy of four mainstream numerical models in different regions and conditions of China during winter.The aim is to provide reference for wind speed forecasting during high-wind periods in winter.Based on four numerical forecast models with different resolutions,initial fields,and assimilation schemes,and utilizing observational data from 131 stations in China,this paper conducts a study and analysis of the error distribution characteristics and forecasting capabilities of wind speed.The study also focuses on typical stations,analyzing the error characteristics and forecasting capabilities in different wind speed ranges and regions.The results show that ensemble forecast models provide more scientifically sound results in complex terrain conditions.High-resolution single models perform well in predicting wind speed variability over simple underlying surfaces.Daytime forecasting is more accurate than nighttime forecasting,and plain areas exhibit the best wind speed forecasting performance.
作者
刘华
马辉
沈晔
郝春宇
俞竣珲
LIU Hua;MA Hui;SHEN Ye;HAO Chunyu;YU Junhui(Shenergy Co.,Ltd.,Shanghai 201103,China;Beijing Goldwind Smart Energy Technology Co.,Ltd.,Beijing 100176,China;Qinghai University,Xining 810016,Qinghai,China)
出处
《电力大数据》
2023年第8期70-78,共9页
Power Systems and Big Data
关键词
数值模式
误差特征
评估算法
风速预报
集合预报
高分辨率
numerical model
error characteristics
evaluation algorithms
wind speed prediction
ensemble prediction
high resolution
作者简介
刘华(1980),男,硕士,高级工程师,主要从事电力企业信息化数字化、网络安全,新能源数字化技术研究及应用等相关工作。
