基于神经网络建模的机床滑动结合面动态特性参数识别  

Dynamic characteristic parameters identification for machine tool sliding joints based on neural network modeling

作　　者：朱坚民[1] 周亚南[1] 何丹丹[1] 郑洲洋 ZHU Jianmin;ZHOU Yanan;HE Dandan;ZHENG Zhouyang(School of Mechanical Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学机械工程学院,上海200093

出　　处：《振动与冲击》2018年第7期109-115,131,共8页Journal of Vibration and Shock

基　　金：国家自然科学基金(50975179);上海市科委科研计划项目(13160502500);沪江基金(D14005)

摘　　要：针对机床滑动结合面动态特性参数难以准确确定的问题。以结合面的刚度参数、阻尼参数为优化变量,建立滑动结合面动态特性参数的神经网络模型,结合神经网络模型的计算结果与机床整机实验模态的分析结果,采用布谷鸟优化算法对结合面的刚度与阻尼参数进行优化识别;以自行设计制造的机床滑动结合面实验台上工作台与床身间的滑动导轨结合面为实例进行了建模、实验、参数识别等分析。分析结果表明:该方法是可行的、有效的,参数识别精度高于已有文献研究。Aiming at the problem that dynamic characteristic parameters of sliding joints of a machine tool are difficult to identify accurately,here,a neural network modeling method for identifying sliding joints’dynamic characteristic parameters was proposed.The model’s variables to be optimized were stiffness parameters and damping ones of sliding joints.With the calculation results of the neural network model and those of the machine tool test modal analysis,stiffness and damping parameter identification of the sliding joints was performed with the cuckoo optimization algorithm.As an example,modeling,tests and parametric identification were conducted for a self-designed and manufactured sliding guide joints between the working platform and the machine tool body.The results showed that the proposed method is feasible and effective;the parametric identification accuracy is higher than those recorded in literatures.

关 键 词：滑动结合面 神经网络建模 动态特性参数 优化识别 实验 

分 类 号：TG502.1[金属学及工艺—金属切削加工及机床]