李晓科, 韩新雨, 赵文波,等.基于概率约束的轴桥车削参数优化[J].河南理工大学学报（自然科学版）,2021,40(1):88-94.

LI X K, HAN X Y, ZHAO W B,et al.Optimization of axle-bridge turning parameters based on probabilistic constraint[J].Journal of Henan Polytechnic University(Natural Science) ,2021,40(1):88-94.

基于概率约束的轴桥车削参数优化

李晓科1, 韩新雨1, 赵文波2, 都金光1,3, 尚宝平1, 肖艳秋1

1.郑州轻工业大学 机电工程学院，河南 郑州  450000;2.洛阳天浩泰轨道装备制造有限公司，河南 洛阳 471000;3.河北省重型智能制造 装备技术创新中心，河北 秦皇岛  066004

摘要:针对轴桥车削时变形大的问题，分别提出相应的工艺改进及加工参数优化策略：通过增加平衡配重减小离心力对车削变形的影响；通过基于概率约束的优化算法选择最优的车削参数，实现车削过程的高效性以及车削精度的可靠性。构建以车削效率为目标的轴桥车削参数优化模型。为充分考虑车削参数的随机变化，以车削变形量小于给定值的概率作为约束。为降低概率约束模型的计算成本，采用基于平移传播的拉丁超立方采样方法确定车削速度、背吃刀量与进给量的参数组合，并通过有限元仿真得出各组参数对应的变形值，进而构建车削变形的克里金模型，在失效概率计算及优化求解过程中直接调用该近似模型。通过对概率约束模型最优点处的参数组合进行试验，验证了本文提出的方法可在轴桥允许变形范围内有效提升车削效率。

关键词:轴桥;概率约束;车削效率;车削参数优化;克里金模型

doi:10.16186/j.cnki.1673-9787.2019120101

基金项目:国家自然科学基金资助项目（51905492 ）；河南省科技攻关项目（192102210069 ）；河南省高等学校重点科研项目 （19A460031）

收稿日期:2019/12/30

修回日期:2020/04/15

出版日期:2021/01/15

Optimization of axle-bridge turning parameters based on probabilistic constraint

LI Xiaoke1, HAN Xinyu1, ZHAO Wenbo2, DU Jinguang1,3, SHANG Baoping1, XIAO Yanqiu1

1.Mechanical and Electrical Engineering Institute， Zhengzhou University of Light Industry， Zhengzhou  450000 ， Henan， China;2.Luoyang TiHot Railway Machinery Manufacturing Co. ， Ltd. ， Luoyang  471000 ， Henan， China;3.Heavy-duty Intelligent Manufacturing Equipment Innovation Center of Hebei Province， Qinhuangdao  066004 ， Hebei， China

Abstract:Aiming at the problem of large deformation during the turning process of the axle-bridge， the corresponding process improvement and processing parameter optimization strategies were proposed respectively ： By introducing the balance weight to reduce the influence of centrifugal force on turning deformation， and by using the optimization algorithm based on probability constraints to select the optimal turning parameters， the efficiency of the turning process and the reliability of turning accuracy were achieved. An optimization model for turning parameters of axle-bridge with the goal of turning efficiency was constructed. In order to fully consider the random variation of turning parameters， the probability that the turning deformation was less than a given value was used as the constraint. In order to reduce the computational cost of the probability constraint model， the Latin hypercube sampling via translational propagation algorithm was used to select the parameter combinations of turning speed， backlash and feed， and the deformation values corresponding to each group of parameters were obtained through finite element simulation. Then， the Kriging approximation model of turning deformation was constructed， which was directly called during the failure probability calculation and optimization solution process. By experimental verification of the optimal parameter combination， the proposed method could effectively improve the turning efficiency within the allowable deformation range of the axle-bridge.

Key words:axle-bridge;probabilistic constraint;turning efficiency;turning parameter optimization;Kriging model

 基于概率约束的轴桥车削参数优化_李晓科.pdf