摘要
变速箱阀体是汽车变速部件中的重要配件。阀体件内部空腔多、深度大,由于其型腔的高度复杂性,传统的有限元法难以对其进行高效精确预测,通常采用反复离线测量和铣削来保证精度。为此,提出了一种基于在机测量的阀体件加工误差实时补偿方法以提高加工精度。首先,通过实验确认主要误差源为装夹变形;其次,基于在机测量获得阀体误差曲面,并利用机器学习算法进行拟合,获得初步补偿模型;进而,针对面铣刀直径大的特点,建立了偏移式补偿轨迹;最后,基于数控机床的坐标原点偏移功能开发实时补偿系统并进行了阀体件加工误差的补偿实验。实验结果表明,补偿后的阀体件加工误差降低55%以上,并具有良好的一致性。
Gearbox valve body is an important part of automobile transmission parts.Because there are many cavities with high complex structure inside the valve body,the traditional finite element method is difficult to predict the internal cavity of the valve body efficiently and accurately.Therefore,this paper proposes a real-time compensation method for machining errors of valve body parts based on on-machine measurement.Firstly,according to the experiment,the main error source of machining deformation is clamping deformation.Secondly,the error surface of valve body was obtained by on-machine measurement,and the machine learning algorithm was used to fit the error surface to obtain the preliminary compensation model.Then,according to the characteristics of large diameter of face milling cutter,the offset compensation track was established.Finally,based on the coordinate origin offset function of CNC machine tool,a real-time compensation system was developed and the machining error compensation experiment of valve body was carried out.The experimental results show that the machining error of the valve body after compensation is reduced by more than 55%,and has good consistency.
作者
韩立彰
葛广言
杜正春
朱梦瑞
陈志祥
HAN Lizhang;GE Guangyan;DU Zhengchun;ZHU Mengrui;CHEN Zhixiang(School of Mechanical Engineering,Shanghai Jiaotong University,Shanghai 200240,China;Shanghai SmartState Technology Co.Ltd.Shanghai 201306,China)
出处
《机械设计与研究》
CSCD
北大核心
2021年第2期78-82,91,共6页
Machine Design And Research
基金
国家自然科学基金资助项目(51975732)
上海市军民融合项目(2016-63)。
关键词
汽车变速箱阀体
装夹变形
在机测量
面铣刀误差补偿
automobile gearbox valve body
clamping deformation
on-machine measurement
error compensation of face milling cutter
作者简介
韩立彰(1991-),男,硕士生,主要研究方向:数控机床误差的测量、建模与补偿;通讯作者:杜正春,E-mail:zcdu@sjtu.edu.cn。
