基于深度神经网络的滑动结合面参数识别研究  

Parameter Identification of Sliding Joint Surface Based on Depth Neural Network

作　　者：刘绪荣 张玮[1] 黄之文[1] 朱坚民[1] Liu Xurong;Zhang Wei;Huang Zhiwen;Zhu Jianmin(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,上海市200093

出　　处：《农业装备与车辆工程》2021年第10期68-72,共5页Agricultural Equipment & Vehicle Engineering

摘　　要：为了解决机床滑动结合面传统建模方法建模精度低和其动态特性参数难确定的问题,提出了一种基于深度神经网络(Deep Neural Network,DNN)的结合面参数识别方法。以M7120D/H卧轴矩台平面磨床的砂轮箱-滑座滑动结合面为研究对象,利用深度神经网络模型预测值和模态分析实验值的相对误差建立目标函数,采用布谷鸟(Cuckoo Search,CS)算法对结合面的刚度、阻尼进行识别研究。结果显示,该方法得到的有限元模型计算值和实验测试值之间的相对误差在3%以内,具有较高的识别精度,表明该方法是正确的、有效的。In order to solve the problems of low modeling accuracy and difficult to determine the dynamic characteristic parameters of machine tool sliding joint surface,a new method based on deep neural network(DNN)is proposed.Taking the sliding joint surface of grinding wheel box sliding seat of M7120D/H horizontal spindle rectangular table surface grinder as the research object,the objective function was established by using the relative error between the predicted value of depth neural network model and the experimental value of modal analysis,and the cuckoo search(CS)algorithm was used to identify the stiffness and damping of the joint surface.The results show that the relative error between the calculated value of the finite element model and the experimental value is less than 3%,and has high recognition accuracy,which shows that the method is correct and effective.

关 键 词：滑动结合面 深度神经网络 拟合模型 参数识别 平面磨床 

分 类 号：TG580[金属学及工艺—金属切削加工及机床]