节点文献

基于多代理模型的自适应约束优化算法及在铆头优化设计中的应用

Self-Adaptive Constraint Optimization Algorithm Based on Multiple Surrogates and its Application in the Design of Rivet Head

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 曲杰胡焱松徐梁马强

【Author】 QU Jie;HU Yansong;XU Liang;MA Qiang;School of Mechanical and Automotive Engineering,South China University of Technology;

【机构】 华南理工大学机械与汽车工程学院

【摘要】 为了求解需要消耗大量计算资源的非线性约束的工程优化问题,提出一种基于多代理模型的自适应约束优化算法.首先给优化问题中的目标函数及每一个约束函数分配一个代理模型候选集,其次通过交叉检验确定每一函数相应候选集内代理模型,对研究问题拟合性能排序,并根据排序结果构造一系列原优化问题的近似模型,应用序列二次规划算法求解,当候选集内代理模型数目不一致时,优先选择具有最优性能的代理模型.候选集内代理模型是保留或删除则基于代理模型拟合性能评价结果确定,新样本通过求解构造的近似模型、非均匀变异算子和混合杂交算子BLX-0. 5三种方式协同获得.最后应用提出的算法求解4个典型的数学优化问题,结果表明基于自适应优化算法得到的近似优化解均较好地逼近于理论最优解;同时应用提出的算法对汽车轮毂轴承单元轴铆工艺中的铆头成形曲面进行优化设计,优化结果较好.

【Abstract】 In order to solve the nonlinear constraint engineering optimization problem which requires high computational resource,a self-adaptive constraint optimization algorithm based on multiple surrogates is proposed. Firstly,every objective function and constraint function is assigned to a candidate surrogate set including multiple interesting surrogates and the ranking of the corresponding surrogates in the each candidate set is determined by the fitting performance of the function on the basis of the cross-validation. Then an approximated model of the original optimization problem develops due to the surrogates which are selected on the basis of the result of the ranking and approximated model,which is solved by the sequential quadratic programming algorithm. When the number of surrogate models is different in every candidate sets for the objective function and constraint functions,the best surrogates are selected preferentially. Whether the surrogates are retained or deleted when the candidate set is determined by the evaluation results of the fitting performance of the function,and the new samples are obtained coordinately by three methods: solving the approximated model,non-uniform mutation operator and blend crossover operator BLX-0. 5.Finally,the proposed constraint optimization algorithm is applied to solve four representative mathematical optimization problems,and the results show that the solutions obtained by the proposed self-adaptive constraint optimization algorithm are better approximating to the theoretical ones of the original optimization problems. At the same time,the proposed method is used to optimize the design of the rivet head forming surface in the shaft clinching process of the vehicle wheel bearing unit,the result is good.

【基金】 广东省科技计划项目(2013B090600019,2012B091100322)~~
  • 【文献出处】 华南理工大学学报(自然科学版) ,Journal of South China University of Technology(Natural Science Edition) , 编辑部邮箱 ,2018年10期
  • 【分类号】O224
  • 【被引频次】2
  • 【下载频次】157
节点文献中: 

本文链接的文献网络图示:

本文的引文网络