摘要
采用混合均匀试验设计法细化铺层参数,确定试验方案,仿真叶片的Tsai-Wu失效因子和最大位移;根据分析结果,构建具有预测能力的径向基神经网络模型,通过计算网络预测值与仿真分析值误差、均方差与均方根值,验证网络模型的可靠性;在完成训练的径向基神经网络中应用遗传算法,径向基神经网络预测值作为遗传算法的输入,对铺层角度、±x°铺层厚度比例和铺层顺序进行迭代求解,获得优化的铺层参数组合。实例表明:优化后叶片的刚度、强度得到改善,验证了方法的可行性和有效性。
The mixed horizontal uniform test design method is used to subdivide the ply parameters and design the test scheme in this paper.The Tsai-Wu failure factor and maximum displacement of blade are simulated,and the radical basis function neural network model with predictive ability is constructed according to the analysis results.The reliability of the RBF neural network is verified by calculating the error,mean square deviation and root mean square value of the network predicted value and the simulation value.The genetic algorithm is integrated into the trained RBF neural network.The predicted value of RBF neural network is used as the input of genetic algorithm.The ply angle,±x°ply thickness ratio and ply stacking sequence of the blade are optimized.Finally,the optimal combination of the ply parameters is obtained.Numerical examples show that the static strength and stiffness of the optimized blade are increased.The feasibility and effectiveness of the method is verified.
作者
赵清鑫
张兰挺
Zhao Qingxin;Zhang Lanting(School of Mechanical Engineering,Inner Mongolia University of Technology,Hohhot 010051,China)
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2020年第4期229-234,共6页
Acta Energiae Solaris Sinica
基金
国家自然科学基金(51665046,51865041)。
关键词
径向基神经网络
遗传算法
结构性能
优化
铺层参数
风力机叶片
radial basis function network
genetic algorithms
structural properties
optimization
ply parameter
wind turbine blade
