复杂下垫面风电场风速数值预测及误差特征研究
投稿时间: 2012-12-24  最后修改时间: 2012-12-24  点此下载全文
引用本文:
摘要点击次数: 442
全文下载次数: 
作者单位E-mail
祖繁 南京信息工程大学大气物理学院 zufan123@yeah.net 
王咏薇 南京信息工程大学大气物理学院  
樊曙先 南京信息工程大学大气物理学院  
吴息 南京信息工程大学大气科学学院  
高卓 北京四方继保自动化股份有限公司  
基金项目:江苏省科技支撑计划;江苏高校优势学科建设工程;长江学者和创新团队发展计划资助
中文摘要:准确的风速预报可供电网及时调度,从而提高风电的经济效益。本文利用WRF模式分别对沿海及山地条件下风电场风速预测进行高分辨数值模拟,并对其误差特征进行分析,结果表明:(1)WRF模式对复杂地形条件下的风速模拟性能良好,模拟值能够较好地体现天气尺度的周期变化;沿海各高度层上模拟效果均明显优于山地;(2)沿海及山地条件下模拟与观测的误差特征各不相同。模式静态数据未能显现沿海的小岛,并且低估了山地测风塔所在的海拔,导致沿海平均模拟风速偏大,山地平均模拟风速偏小。(3)分析不同风向的归一化均方根误差,沿海条件下,陆风情况下,下垫面相对复杂,误差明显增大;海风情况下,下垫面均一,误差明显减小。该种误差分布特征为沿海风电场风速预报误差订正提供了方向。(4)仅作单个风电场周边数百平方公里的预测,采用一台12核的服务器进行WRF模式的并行计算可满足48h短期预测的时效性。然而仅仅提高模拟的网格分辨率,并不一定能提升模拟的准确性,还需要从引入更准确的地形地貌数据入手提高模拟的精度。
中文关键词:复杂下垫面  WRF模式  风速预测  误差特征  网格分辨率
 
Numerical Simulation and Error Characteristics for Wind Speed Forecasting on Wind Farm over Complex Underlying Surface
Abstract:The accurate wind speed forecasts ensure the power dispatching timely, thereby improving the economic benefits of wind power. Wind speeds on wind farm in coastal and mountain area are respectively simulated using WRF model and the error characteristics of wind speed forecasting are discussed in this article. The results indicate that: (1) WRF model has a good performance in simulating the wind speed over complex underlying surface, and the synoptic scale variations of wind speed are reflected in the simulation. The simulation results in coastal area are superior to those in mountain area in each layer. (2) The error characteristics between simulations and observations are different in coastal and mountain area. The small island is not appeared in coastal area and the elevation where the wind tower located is underestimated in mountain area in the static terrestrial data. As a result, the average simulated wind speeds are overestimated in coastal area and underestimated in mountain area. (3) The normalized root mean square error of wind speeds are analyzed in different directions. The error is significantly increased due to the relative complex underlying surface in the case of land breezes and decreased due to the homogeneous underlying surface in the case of sea breezes in coastal area. The obvious error distribution characteristics provide a direction for further error correction of wind speed forecasting on wind farm in coastal area. (4) To forecast the wind speed on a scale of few hundred kilometers surrounding the wind farm, parallel computation of WRF model using a server with 12-core processors is sufficient to meet the time effectiveness of 48h short term forecasts. However, only increasing the grid resolution is not necessary to improve the accuracy of the simulation, the more accurate terrestrial data should be introduced to improve the precision.
keywords:complex underlying surface  WRF model  wind speed forecasting  error characteristics  grid resolution
查看全文  查看/发表评论  下载PDF阅读器