摘要
建立人工神经网络、径向基函数网络和支持向量回归机三种近似模型,结合蒙特卡洛方法与表征粗糙度参数随机特性的概率模型对风力机翼型气动性能进行不确定性分析。结果表明,支持向量回归机具有最佳预测精度。对于风力机翼型FX 63-137,最大升力系数对吸力面前缘粗糙度的敏感性明显高于压力面;对于吸力面或压力面,前缘粗糙带厚度对最大升力系数的影响稍大于其覆盖长度的影响。研究工作为风力机翼型的鲁棒性设计优化奠定了理论基础。
Three metamodels.i.e.,artificial neural network（ANN）,radial basis function（RBF） and support vector regression（SVR） have been constructed.With the combination of the metamodel method,Monte Carlo method and the probability distributions of roughness parameters,the uncertainty analysis of the aerodynamic performance of the wind turbine airfoil has been conducted. The results show that the SVR performs the best in model approximation.For the airfoil FX 63-137,the maximum lift coefficient is much more sensitive to the surface roughness of the suction side（SS） than that of the pressure side（PS）;for either SS or PS,the influence of the roughness height on the aerodynamic performance is slightly larger than that of the roughness covering length. This study can provide theoretical foundation for the robust optimization design of wind turbine airfoil.
出处
《工程热物理学报》
EI
CAS
CSCD
北大核心
2012年第5期784-787,共4页
Journal of Engineering Thermophysics
基金
国家自然科学基金项目(No.51176146)
国家科技支撑计划课题(No.2012BAA08B06)
关键词
近似模型
粗糙度
风力机翼型
气动性能
蒙特卡洛方法
metamodel
roughness
wind turbine airfoil
aerodynamic performance
Monte Carlo method
