首页|基于多目标优化的箱梁渡槽结构设计研究

基于多目标优化的箱梁渡槽结构设计研究

扫码查看
针对箱梁渡槽所受水荷载远大于同等跨径箱梁桥所受汽车荷载导致槽身纵向受力突出的问题,在渡槽结构设计中统筹考虑箱梁纵向应力储备和造价,构建了以截面尺寸、钢筋用量等6个参数为设计变量,以渡槽边缘正应力和造价为目标函数,以槽身挠度、法向应力等9个验算指标为约束条件的多目标优化模型,并基于支持向量机近似模型和NSGA2遗传算法求解多目标优化模型的Pareto最优解集,然后综合专家打分、COWA算子、博弈论综合赋权与TOPSIS法从Pareto最优解集中选出最佳方案.结果表明:与初始方案相比,最佳方案的渡槽边缘正应力储备提高了 17.55%,渡槽上部结构总造价降低8.84%;与有限元分析相比,该方法在保持较高计算精度的同时节省了 65.42%的计算时间成本.
Research on structural design of box girder aqueduct based on multi-objective optimization
Aiming at the problem that the water load on the box girder aqueduct is much larger than the vehicle load on the same span box girder bridge,the longitudinal stress of the aqueduct body is prominent,in the design of the aqueduct structure,the longitudinal stress reserve and cost of the box girder are considered as a whole,a multi-objective optimization model is constructed with six parameters such as section size and steel consumption as design variables,the normal stress and cost of the aqueduct edge as objective functions,and nine checking indexes such as the deflection and normal stress of the aqueduct body as constraints.Based on the support vector machine approximation model and NSGA2 genetic algorithm,the Pareto optimal solution set of the multi-objective optimization model is solved.Then,the best solution is selected from the Pareto optimal solution set by combining expert scoring,COWA operator,game theory comprehensive weighting and TOPSIS method.The results show that compared with the initial scheme,the normal stress reserve of the aqueduct edge of the optimal scheme is increased by 17.55%,and the total cost of the aqueduct superstructure is reduced by 8.84%.Compared with the finite element analysis,this method saves 65.42%of the calculation time cost while maintaining high calculation accuracy.

box girder aqueductmulti-objective optimizationstructure designapproximate modelNSGA2 genetic algorithm

赖成联、刘潇、刘惠娟、祝通华、谢开仲

展开 >

广西水利电力勘测设计研究院有限责任公司,广西南宁 530023

中国水电基础局有限公司,天津 301700

广西水利电力职业技术学院土木建筑工程学院,广西南宁 530023

广西大学土木建筑工程学院,广西南宁 530004

展开 >

箱梁渡槽 多目标优化 结构设计 近似模型 NSGA2遗传算法

国家自然科学基金广西壮族自治区重点研发计划

51868007桂科AB22036007

2024

广西大学学报(自然科学版)
广西大学

广西大学学报(自然科学版)

CSTPCD北大核心
影响因子:0.767
ISSN:1001-7445
年,卷(期):2024.49(2)
  • 28