China Mechanical Engineering ›› 2021, Vol. 32 ›› Issue (13): 1555-1561,1570.DOI: 10.3969/j.issn.1004-132X.2021.13.006

Previous Articles     Next Articles

Multi-objective Optimization of Cutting Parameters in Ti6Al4V Milling Processes Based on Finite Element Simulation#br#

LI Yujia1;HUANG Bing1;LU Juan1,2;ZHONG Qijing1;CHEN Chaoyi1;LIAO Xiaoping1;MA Junyan1   

  1. 1.Guangxi Key Laboratory of Manufacturing Systems and Advanced Manufacturing Technology,Guangxi University,Nanning,530004
    2.Department of Mechanical and Marine Engineering,Beibu Gulf University,Qinzhou,Guangxi,535011
  • Online:2021-07-10 Published:2021-07-16

基于有限元模拟的Ti6Al4V铣削过程参数多目标优化

黎宇嘉1;黄兵1;鲁娟1,2;钟奇憬1;陈超逸1;廖小平1;马俊燕1   

  1. 1.广西大学广西制造系统及先进制造技术重点实验室,南宁,530004
    2.北部湾大学机械与船舶海洋工程学院,钦州,535011
  • 通讯作者: 鲁娟(通信作者),女,1988年生,副教授。研究方向为智能制造与模式识别。E-mail: lujuan3623366@163.com。
  • 作者简介:黎宇嘉,男,1995年生,硕士研究生。研究方向为智能制造与仿真。E-mail:535025526@qq.com。
  • 基金资助:
    国家自然科学基金 (51665005);
    广西自然科学基金重点项目(2020JJD160004);
    广西自然科学基金(2019JJB160048);
    广西高校中青年教师科研基础能力提升项目(2020KY10014)

Abstract: In order to make the cutting processes meet the requirements of environmentally conscious manufacturing(ECM), for the quality index(surface roughness) and ECM index(energy consumption), aiming at the Ti6Al4V processes, the Gaussian process regression(GPR) method optimized by artificial bee colony (ABC) algorithm was used to established the finite element agent model, and the processing parameters satisfying the optimal machining objectives were obtained by using the multi-objective particle swarm optimization(MOPSO) algorithm. Deform-3D, a finite element simulation software which might reduce the testing cost was used to obtain the surface roughness and energy consumption data corresponding to each milling parameter combinations, and the effectiveness was proved by physical tests. Then, an improved GPR method was used to establish the prediction model based on the surface roughness and energy consumption of the finite element simulation data. The performance of the model was compared with that of the other two models, and the advantages of the improved model in accuracy and response time were proved. Finally, Pareto front of processing parameters with the goal of minimum energy consumption and excellent surface quality were obtained by MOPSO algorithm. The effectiveness of the ABC-GPR-MOPSO algorithm was verified by physical tests.

Key words:  , milling process, finite element simulation, Gaussian process, Pareto front

摘要: 为使切削加工过程满足环境意识制造(ECM)的要求,针对质量指标(表面粗糙度)和ECM指标(能耗),针对Ti6Al4V的铣削过程,采用人工蜂群(ABC)算法优化的高斯过程回归(GPR)方法构建有限元代理模型,并采用多目标粒子群优化(MOPSO)算法获得满足最优加工目标的加工参数。为减少试验成本,采用有限元仿真软件Deform-3D获取各铣削参数组合对应的表面粗糙度和能耗数据,并通过物理试验验证其有效性;基于仿真数据,利用改进的GPR方法构建预测表面粗糙度和能耗的代理模型,并对比了该模型与其他两种模型的性能,证明了改进模型在精度和响应时间上的优势;采用MOPSO算法,以最小能耗和优良表面质量为目标,优化得到加工参数的Pareto前沿,并用物理试验验证了ABC-GPR-MOPSO算法的有效性。

关键词: 铣削加工, 有限元模拟, 高斯过程, Pareto前沿

CLC Number: