基于GPU的结构静力拓扑优化设计方法  

GPU-Based Structural Static Topology Optimization Design Method

作　　者：吴超 WU Chao(School of Automotive and Mechanical Engineering,Changsha University of Science and Technology,Changsha 410014,China)

机构地区：[1]长沙理工大学汽车与机械工程学院,湖南长沙410014

出　　处：《河南科技》2022年第10期11-15,共5页Henan Science and Technology

基　　金：国家自然科学基金资助项目“基于静强度和动力学要求的双模量材料结构拓扑和参数优化方法研究”(12171065);中国建筑第五工程局有限公司科研项目“基于AI的多工况条件下建筑结构实用有限元分析方法研究及其分析平台开发”(3030402)。

摘　　要：针对连续体结构拓扑优化存在的计算量大、计算效率低等问题,开展了基于GPU并行计算的大规模结构静力拓扑优化方法研究。首先,为了减少有限元分析的迭代次数,引入了雅可比(Jacobi)对角线预处理器,研究基于共轭梯度法和预处理技术的结构有限元并行计算方法。其次,基于单元免组装技术,结合并行迭代计算方法,研究基于GPU的结构静力拓扑优化并行计算方法。在完成上述方法的Matlab和C++并行计算核函数编程后,进行了大量的算例考核。通过给出的算例来验证提出方法的有效性和计算效率,结果表明,该方法具有重要的理论价值和工程应用前景。Aiming at the problems of large amount of calculation and low efficiency in topology optimization of continuum structures,the static topology optimization method of large-scale structures based on GPU parallel computing is studied.Firstly,in order to reduce the number of iterations of finite element analysis,Jacobi diagonal preprocessor are introduced to study the parallel calculation method of structural finite element based on conjugate gradient method and preprocessing technology.Secondly,based on the element free assembly technology and the parallel iterative calculation method,the parallel calculation method of structural static topology optimization based on GPU is studied.Completed the Matlab and C++parallel computing kernel function programming of the above method,and completed a large number of numerical examples.Numerical examples are given to verify the effectiveness and computational efficiency of the proposed method,and show that the proposed method has important theoretical value and engineering application prospect.

关 键 词：拓扑优化 GPU并行 免组装方法 共轭梯度法 预处理器 

分 类 号：TB34[一般工业技术—材料科学与工程]