Application of an Improved Lagrangian Multiplier Solving Method in Topology Optimization
Aiming at the shortages of continuum topology optimization of the existing Lagrangian multiplier algorithms, an improved Lagrangian multiplier solving method is discussed.The paper introduces the topology optimization mathematical model of variable density material interpolation and describes the derivation process of optimal criteria formula based on the mentioned mathematical model.The two main existing Lagrangian multiplier algorithms in optimal criteria formula, the numerical method and analytical method, are introduced.After the characteristics of the two algorithms in volume constraints have been theoretical analyzed, through absorbing advantages of the two algorithms, an improved Lagrangian multiplier solving method is proposed.The improved algorithm is applied to structural optimize the MBB beam in combination with ANSYS by APDL language, and contrasted with the original topology optimization tools in ANSYS.Through the further development, two possible situations caused by wrong initial settings are introduced.It is verified that losing information as a result of one-order Taylor expansion will cause excessive volume error in analytical method and the improved Lagrangian multiplier solving method has high engineering application.
Topology Optimization Variable Density Method Optimal Criteria Lagrangian Multiplier Solving Method
Wang Yong Sun Boyu Li Kun Zhang Teng
School of Mechanical and Automobile Engineering,Hefei University of Technology,230009,Hefei,China
国际会议
黄山
英文
1-7
2012-06-18（万方平台首次上网日期，不代表论文的发表时间）