基于并行计算和遗传算法的钢-UHPC华夫板组合梁优化设计  

Optimal design of steel-UHPC composite beam with waffle slab using parallel computing and genetic algorithm

作　　者：史腾 朱劲松[1,2] 王子挺 秦亚婷 SHI Teng;ZHU Jin-song;WANG Zi-ting;QIN Ya-ting(School of Civil Engineering,Tianjin University,Tianjin 300072,China;Key Laboratory of Coast Civil Structure Safety of Ministry of Education,Tianjin University,Tianjin 300072,China)

机构地区：[1]天津大学建筑工程学院,天津300072 [2]天津大学,滨海土木工程结构与安全教育部重点实验室,天津300072

出　　处：《计算力学学报》2023年第3期357-365,共9页Chinese Journal of Computational Mechanics

基　　金：国家自然科学基金(52078333);天津市交通运输科技计划(2019B-21)资助项目。

摘　　要：为实现钢-超高性能混凝土(UHPC)华夫板组合梁结构快速经济合理的设计,提出了基于并行计算与遗传算法的结构优化设计方法.通过Python建立了并行计算平台,使Abaqus和Python能够执行同步数值模拟和数据处理,以成本最小化为目标,采用遗传算法对钢-UHPC华夫板组合梁进行了优化,验证了所提方法的可行性.结果表明,遗传算法中密集的分析任务可以并行化并分配给不同的计算资源以提高计算效率;使用并行计算可以提高8.6倍的优化效率;并行计算和串行计算的CPU平均使用率分别为82%和18%.本文方法的成功应用可为其他类型结构的优化设计提供参考.A structural optimal design method based on parallel computing and genetic algorithm is proposed to realize the rapid,economical,and reasonable design scheme for structures of steel-UHPC composite beam with waffle slabs.A parallel computing platform was established in Python to enable Abaqus and Python to perform the simultaneous numerical simulation and data processing.The genetic algorithm was used to optimize the steel-UHPC composite beams with waffle slabs with the goal of minimizing cost,which verified the feasibility of the proposed method.The results show that the intensive computational tasks in the genetic algorithm can be parallelized and distributed to different computational resources to improve computing efficiency;the application of parallel computing can increase the optimization efficiency by 8.6 times;the average CPU usage of parallel computing and sequential computing is 82% and 18%,respectively.The successful application of the proposed method can provide a reference for the optimal design of other types of structures.

关 键 词：超高性能混凝土 华夫板 组合梁 结构优化设计 并行计算 遗传算法 

分 类 号：TU398.9[建筑科学—结构工程] O224[理学—运筹学与控制论]