考虑地形影响的短期风电功率预测  

Short-term wind power prediction methods considering the influence of terrain

作　　者：赵川 陈根军[2] 叶华 祝明乐 吴枭[3] 

机构地区：[1]云南电力调度控制中心,云南昆明650011 [2]南京南瑞继保电气有限公司,江苏南京211102 [3]东南大学电气工程学院,江苏南京210096

出　　处：《应用科技》2015年第6期6-9,共4页Applied Science and Technology

基　　金：国家科技支撑计划重大项目(2013BAA01B00);国家自然科学基金资助项目(51377021)

摘　　要：风资源具有很强的随机性和间歇性,随着大量的风电功率并网,势必会应影响电力系统的安全、稳定运行,降低电能质量。首先用BP神经网络预测出测风塔处的风速,再进一步考虑地形因素的影响,用CFD软件对风电场风流进行数值模拟,计算出各风机轮毂高度处的风加速因数和水平偏差等数据,然后用MATLAB软件编程求出各风机轮毂高度处的风速,再根据风力发电机的功率曲线算出预测功率。提出了考虑地形影响的短期风电功率预测方法,预测效果较为理想,适合实际工程应用。Wind is a resource with strong randomness and intermittence. With a lot of wind power joining the grid,it is bound to endanger the security and stability of the power system. Besides,it may worsen power quality. First,this paper predicted the wind speed by the method of BP neural network according to historical data,and used CFD software to simulate the numerical operation of the farm Merry when further taking the impact of terrain into consideration,deriving the wind acceleration factor and the level bias and other data at each fan hub height. Second,the wind speed of each fan hub height was calculated by MATLAB software programming. Finally,the predicted power was estimated according to the power curve of the wind turbine,and thereby the short-term wind power prediction methods considering the influence of terrain was proposed. This model proposed in this paper has relatively higher forecasting accuracy,which is suitable for engineering application.

关 键 词：风电功率 预测 BP神经网络 CFD软件 MATLAB 

分 类 号：TK8[动力工程及工程热物理—流体机械及工程]