首页|基于屈曲承载能力的索支撑柱撑杆长度优化方法研究

基于屈曲承载能力的索支撑柱撑杆长度优化方法研究

扫码查看
索支撑柱是一种适用于大跨度空间结构支撑的新型结构构件,撑杆长度是影响其屈曲承载能力的关键设计参数.分别基于梯度法和多种群遗传算法设计两种索支撑柱撑杆长度优化方法,并对两种优化结果进行分析和对比.通过推导索支撑柱线性屈曲荷载的梯度计算公式,阐述多种群遗传算法的重要策略和参数取值,分别确定出两种优化算法的优化流程,结合优化流程对单层索支撑柱和三层索支撑柱进行优化.结果表明,基于梯度法的优化方法计算效率高、收敛速度快,但对于多变量优化问题容易陷入局部最优解;而多种群遗传算法依赖大量有限元分析,但可以使用并行计算,在保证优化效率的同时,对于多变量优化问题能够有效收敛至全局最优解.研究成果可运用于大跨度屋盖支撑结构中,能够更为有效地提高建筑结构材料利用率,减少建材用量.
Research on Crossarm Length Optimization Method of Prestressed Stayed Column Based on Buckling Capacity
Prestressed stayed column is a new type of structural component which suitable for large-span spatial structure support.The crossarm length of the prestressed stayed column is a key design parameter that affects its buckling capacity.This article designed two optimization methods for the crossarm length of prestressed stayed columns based on gradient algorithm and multi-population genetic algorithm,and compared the optimization results of the two methods.By deriving the gradient calculation formula for linear buckling load of prestressed stayed columns and elaborating the important strategies and parameters of multi-population genetic algorithms,the optimization processes of two optimization algorithms were determined.And the optimization process is combined to optimize the single-layer cable support column and the three-layer cable support column.The results show that the optimization method based on gradient method has high computational efficiency and fast convergence speed,but it is prone to fall into in local optimal solutions for multivariate optimization.However,multi-population genetic algorithm relies on finite element analysis,but can use parallel computing to ensure optimization efficiency while effectively converging to the global optimal solution for multivariate optimization problems.The research results can be applied to large-span structures,which can effectively improve the utilization rate of structural materials and reduce the amount of building materials used.

prestressed stayed columnstructural optimizationgradient algorithmmulti-population genetic algorithmlinear buckling load

杨位珂

展开 >

中铁十四局集团建筑工程有限公司 山东济南 250014

索支撑柱 结构优化 梯度法 多种群遗传算法 线性屈曲荷载

2024

铁道建筑技术
中国铁道建筑总公司

铁道建筑技术

影响因子:0.539
ISSN:1009-4539
年,卷(期):2024.(10)
  • 1