复合材料科学与工程 ›› 2022, Vol. 0 ›› Issue (6): 5-9.DOI: 10.19936/j.cnki.2096-8000.20220628.001

• 基础研究 •    下一篇

幂函数过滤函数不同参数对层合板拓扑优化收敛率的影响

闫金顺1,2,3, 孙鹏文3*, 马志坤3, 赵雄翔4, 董新洪3   

  1. 1.内蒙古工业大学 能源与动力工程学院,呼和浩特 010051;
    2.吕梁学院 矿业工程系,吕梁 033001;
    3.内蒙古工业大学 机械工程学院,呼和浩特 010051;
    4.内蒙古自治区能源技术中心,呼和浩特 010010
  • 收稿日期:2022-03-14 出版日期:2022-06-28 发布日期:2022-07-19
  • 通讯作者: 孙鹏文(1966-),男,博士,教授,主要从事结构设计与优化方面的研究,pwsun@imut.edu.cn。
  • 作者简介:闫金顺(1986-),男,博士研究生,讲师,主要从事结构设计与优化方面的研究。
  • 基金资助:
    国家自然科学基金(52165035,51865041);内蒙古自治区研究生科研创新资助项目(BZ2020037,S20210186Z)

Influence of different parameters of filter function in the form of power function on the convergence rate of laminate topology optimization

YAN Jin-shun1,2,3, SUN Peng-wen3*, MA Zhi-kun3, ZHAO Xiong-xiang4, DONG Xin-hong3   

  1. 1. College of Energy and Power Engineering, Inner Mongolia University of Technology, Hohhot 010051, China;
    2. Department of Mining Engineering, Luliang University, Luliang 033001, China;
    3. School of Mechanical Engineering, Inner Mongolia University of Technology, Hohhot 010051, China;
    4. Energy Technology Center of Inner Mongolia Autonomous Region, Hohhot 010010, China
  • Received:2022-03-14 Online:2022-06-28 Published:2022-07-19

摘要: 论文探究了过滤函数在基于独立连续映射法复合材料层合板拓扑优化数学模型建立和求解中的作用以及幂函数形式过滤函数不同参数对优化收敛率的影响,引入幂函数形式过滤函数,建立了位移约束下以质量最小为目标函数的拓扑优化数学模型,推导了目标函数和约束函数的灵敏度计算公式,采用序列二次规划算法求解。结果表明:幂函数形式过滤函数的参数不同,优化收敛率亦不同,选择合适的参数可提高收敛率。同时,提高收敛率可降低映射反演结果对给定替换阈值的依赖程度,减少连续解与离散解目标函数的差值,提升结果的可靠性。

关键词: 过滤函数, 收敛率, 独立连续映射法, 拓扑优化, 复合材料层合板

Abstract: The role of the filter function in the establishment and solution of the mathematical model for composite laminates topology optimization based on independent continuous mapping method, and the influence of different parameters of the filter function in the form of power function on the optimization convergence rate is studied in this paper. Through introducing the filter function in the form of power function, the mathematical model of topology optimization with the minimum mass as the objective function under displacement constraint is established, the sensitivity formulas of the objective function and the constraint function is deduced, and the sequential quadratic programming algorithm is used to solve. The results show that the optimal convergence rate varies with the parameters of the filter function in the form of power function, and choosing appropriate parameters can improve the convergence rate. At the same time, improving the convergence rate can reduce the dependence of the mapping inversion result on the given replacement threshold, reduce the difference between the objective function of the continuous solution and the discrete solution, and increase the result reliability.

Key words: filter function, convergence rate, ICM method, topology optimization, composite laminate

中图分类号: