GRAPES模式在华南沿海风电场精细风场预报中的应用  

Application of the GRAPES model to wind field fine forecasting in wind farms in southern coastal China

作　　者：蒋德海[1] 宋耀明[2] 陈子通[1] 

机构地区：[1]中国气象局广州热带海洋气象研究所,广东省区域数值天气预报重点实验室,广州510080 [2]气象灾害教育部重点实验室(南京信息工程大学),南京210044

出　　处：《资源科学》2015年第2期325-332,共8页Resources Science

基　　金：国家自然科学基金资助项目(编号:41105006);广东科技计划项目(编号:2012A032300013;2012A061400012);广东省气象局科技项目(编号:2013B09);广东省自然科学基金博士启动项目(编号:S2011040002251);公益性行业(气象)科研专项项目(编号:GYHY201006035)

摘　　要：将GRAPES模式通过嵌套降尺度、垂直网格加密等精细化处理手段和边界层参数化方案的优选,为华南沿海风电场提供精细风场预报。气象模式分别采用6km×6km和1km×1km水平分辨率,并配置不同分辨率的地形和下垫面资料。边界层参数化方案分别采用GRAPES自带的MRF方案和MYJ方案针对不同天气类型进行预报。将预报结果与测风塔同期观测资料进行比较评估,结果表明:模式分辨率的提高并结合较精细的地形和下垫面资料,能有效提高风速的预报精度;通过精细化处理,GRAPES模式对风力发电的各风速段预报效果都有明显改善;模式边界层参数化方案的选取与地形、主导风向和大气稳定度的配置密切相关,通过对天气形势的预判断和集合预报结果的后处理,能够有效提高风速的预报精度。Wind power construction projects have commenced in coastal areas of southern China in order to alleviate energy contradictions. Investigation and evaluation of wind energy resources and daily weather forecast services for wind power are required to provide detailed and accurate wind field numerical forecasting. A good forecast effect needs targeted optimization of the local physical process and parameterization scheme. The GRAPES model is used to provide the wind farms of coastal southern China fine forecasts of wind fields by nested downscaling,vertical grid refinement,and selective planetary boundary layer parameterization schemes. Horizontal resolution are 6km × 6km and 1km × 1km in the meteorological model,and are configured with different resolution topographical data and underlying surface data. Two planetary boundary layer parameterization schemes,MRF and MYJ,are used to forecast different weather types. Comparing simulations and data from anemometers,we found that refinement of the GRAPES model has a great influence on the wind field forecast. The effective improvement of wind speed simulation needs synchronization improvement of model resolution,terrain and underlying surface data resolution because of the influence of topography and underlying surface smoothness or interpolation degree. Higher model resolution and refined topographical and underlying surface data can improve the accuracy of wind speed forecasting. The GRAPES forecasting effect on variable speed of wind power generation has been obviously enhanced by refinement. The selection of planetary boundary layer parameterization schemes is closely related to terrain,prevailing wind direction,and atmospheric s Tableilities. By pre-estimating synoptic situations and reprocessing ensemble predictions,the accuracy of wind speed forecasting has been improved by8% and 12% respectively.

关 键 词：风电场 GRAPES模式 精细化 边界层参数化方案 华南沿海 

分 类 号：P457.5[天文地球—大气科学及气象学]