首页|基于改进Kriging模型的风力机翼型气动优化设计

基于改进Kriging模型的风力机翼型气动优化设计

扫码查看
针对目前风力机翼型气动性能的优化需求,提出一种基于改进Kriging模型的风力机翼型气动优化设计方法.首先,采用CST翼型参数化方法,将初始翼型的几何形状转换为可供计算机识别的参数化模型;随后,通过最小二乘法原理的加权处理,建立含有加权变异函数的改进Kriging模型,与小生境遗传算法相结合,并从参数化模型的变量空间中选取适量样本构建翼型,以阻力系数及力矩系数最小化为目标,以升力系数及最大厚度不低于初始值为约束条件进行翼型的气动优化设计;最后,通过CFD模拟进行优化翼型的仿真验证,并在相同的来流攻角下,将优化翼型与初始翼型进行气动特性对比.结果表明:优化翼型的阻力系数、力矩系数分别降低0.42%,0.25%,同时升力系数增加6.25%,证明了该方法有效性.这为风力机翼型气动设计提供了一种新途径.
Aerodynamic Optimization Design of Wind Turbine Airfoils Based on Improved Kriging Model
An aerodynamic optimization design method based on improved Kriging model for wind turbine airfoils is proposed to meet the current aerodynamic performance optimization requirements.Firstly,the CST airfoil parameterization method is used to convert the initial airfoil geometry into a parameterized model which can be identified by a computer.Then,an improved Kriging model with weighted variogram is established by the weighted processing of the least squares principle.Combined with niche genetic algorithm,the airfoil is constructed by selecting appropriate samples from the variable space of the parameterized model,aiming at minimizing the drag coefficient and moment coefficient.The aerodynamic optimization design of airfoils is carried out under the constraints of lift coefficient and maximum thickness not lower than the initial value.Finally,the optimization airfoils are verified by CFD simulation,and at the same angle of attack of incoming flow,the aerodynamic characteristics of the optimized airfoil are compared with that of the initial airfoil.Results show that the drag coefficient and moment coefficient are reduced by 0.42%and 0.25%respectively,at the same time,the lift coefficient is increased by 6.25%.The validity of the method is proved,which provides a new approach to aerodynamic design of wind turbine airfoils.

least squares methodKriging modeloptimization designairfoil parameterization

孟祥恒、文泽军、肖钊、张帆

展开 >

昌吉学院能源与控制工程学院,新疆昌吉 831100

湖南科技大学机电工程学院,湖南湘潭 411201

最小二乘法 Kriging模型 优化设计 翼型参数化

国家自然科学基金资助项目国家重点研发计划湖南省自然科学基金资助项目

519051652016YFF02034002022JJ90003

2024

湖南科技大学学报(自然科学版)
湖南科技大学

湖南科技大学学报(自然科学版)

CSTPCD北大核心
影响因子:0.675
ISSN:1672-9102
年,卷(期):2024.39(3)