中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2023, Vol. 42 ›› Issue (6): 9-17.DOI: 10.3969/j.issn.1674-0696.2023.06.02

• 交通+大数据人工智能 • 上一篇    

基于遗传算法的斜拉桥成桥索力优化应用研究

王涛1,胡宇鹏2,张兴标1,王路1   

  1. (1. 西南科技大学 土木工程与建筑学院,四川 绵阳 621000; 2. 中国工程物理研究院总体工程研究所,四川 绵阳 621999)
  • 收稿日期:2021-10-08 修回日期:2022-08-18 发布日期:2023-08-01
  • 作者简介:王 涛(1983—),男,四川南充人,副教授,博士,主要从事桥梁工程方面的研究。E-mail:7015294@qq.com
  • 基金资助:
    国家自然科学基金项目(51708468);西南科技大学博士基金项目(20zx7125)

Application of Completed Cable Force Optimization of Cable-Stayed Bridge Based on Genetic Algorithm

WANG Tao1, HU Yupeng2, ZHANG Xingbiao1, WANG Lu1   

  1. ( 1. School of Civil Engineering and Architecture, Southwest University of Science and Technology, Mianyang 621000, Sichuan, China; 2. Institute of System Engineering, China Academy of Engineering Physics, Mianyang 621999, Sichuan, China)
  • Received:2021-10-08 Revised:2022-08-18 Published:2023-08-01

摘要: 为研究智能优化算法在斜拉桥索力优化中的应用,将遗传算法与有限元算法进行嵌入式融合,以自主开发的有限元程序为基础,使用杆、梁单元构建大跨度公铁两用斜拉桥全桥三维有限元计算模型,讨论了索力对于各个单元节点弯矩的影响矩阵计算方法。建立了2种索力优化目标函数,使用MATLAB集成的遗传算法工具箱进行求解,同时给出了可以直接应用的计算代码,探讨不同目标函数的优化计算结果差别;使用多项式拟合方法对索力优化结果进行调整,讨论了考虑、不考虑几何非线性效应的成桥状态计算结果差别。研究结果表明:使用基于遗传算法的智能随机搜索可有效地求解与索力相关的目标函数极值,辅助桥梁设计人员得到最优的斜拉桥成桥索力。

关键词: 桥梁工程;斜拉桥;遗传算法;斜拉索;索力优化

Abstract: In order to study the application of intelligent optimization algorithm in cable force optimization of cable-stayed bridge, genetic algorithm and finite element algorithm were embedded and combined. Based on the self-developed finite element program, a 3D FEM calculation model of the whole large-span cable-stayed bridge for both highway and railway use was established by using rod and beam elements, and the calculation methods of the influence matrix of cable force on the bending moment of each element node were discussed. Two kinds of cable force optimization objective functions were established and solved by using the genetic algorithm toolbox integrated by MATLAB. Meanwhile, the calculation codes which could be directly applied were given, and the difference of optimization results of different objective functions was discussed. Polynomial fitting method was used to adjust the cable force optimization results, and the difference between the calculation results of bridge state with and without considering geometrical nonlinear effect was discussed. The research results show that the intelligent random search based on genetic algorithm can effectively solve the extremum of the objective function related to cable forces and assist bridge designers to obtain the optimal cable forces of cable-stayed bridges.

Key words: bridge engineering; cable-stayed bridge; genetic algorithm; stay-cables; cable force optimization

中图分类号: