摘要
针对各向同性虚拟材料模型不能准确表征螺栓固定结合面法向和切向的性能差异的问题,以机床立柱-床身螺栓固定结合面为研究对象,采用横观各向同性虚拟材料对机床螺栓固定结合面进行动力学建模,通过试验设计和有限元分析构建机床螺栓固定结合面的径向基函数神经网络模型。结合模态试验结果构建寻优目标函数,并利用粒子群优化算法进行求解,识别出横观各向同性虚拟材料弹性常数。结果表明:实验模型的理论模态与实验模态前六阶振型基本一致,固有频率的误差小于5.1%,说明了该方法的有效性,为机床螺栓固定结合面参数识别提供了一种新方法。
Aiming at the problem that isotropic virtual material model can′t accurately characterize the difference of normal and tangential performances of bolt-fixed joint surface,in this paper,the dynamic model of bolt-fixed joint surface of machine tool is built with transversely isotropic virtual material,and based on the experimental design and finite element analysis;and a radial basis function(RBF)neural network model of bolt-fixed joint surface of machine tool is made.Based on the modal test results,the optimization objective function is constructed and solved by particle swarm optimization(PSO),the elastic constants of transversely isotropic virtual materials are identified.The results show that the first six vibration modes of the theoretical model basically coincide with the experimental modes respectively,and the error of natural frequency is less than 5.1%.This study shows the effectiveness of the method,and provides a new method for identifying parameters of the fixed joint surface of machine tool bolts.
作者
李威
黄晓华
邢炜烽
赵吉庆
李坤鹏
Li Wei;Huang Xiaohua;Xing Weifeng;Zhao Jiqing;Li Kunpeng(School of Mechanical Engineering,Nanjing University of Technology,Nanjing 210094,China)
出处
《机械科学与技术》
CSCD
北大核心
2020年第1期16-21,共6页
Mechanical Science and Technology for Aerospace Engineering
基金
国家科技重大专项项目(2013ZX04002-011)资助.
关键词
机床螺栓固定结合面
横观各向同性虚拟材料
径向基函数神经网络
粒子群优化算法
bolt fixed joint of machine tool
transversely isotropic virtual material
RBF neural network
PSO
vibration mode
dynamic parameter identification
modal test
