摘要
为充分发挥钢框架-钢板剪力墙结构的抗侧移能力和承载能力,实现高层建筑要求的双重抗侧力体系和多道抗震设防的要求,需要设计出一个拥有合理破坏模式和耗能良好的结构.传统设计思路多要进行多次人工假定和修正,即使优化也多是针对单一的性能目标或是单一的方向进行,加上人工假定和修正的效率低下,难以在较短时间内获得符合结构多性能目标的全局最优解.文中从结构的优化设计角度出发,以结构的性能为目标,在钢板剪力墙简化计算模型基础上,将钢框架-钢板剪力墙的塑性设计问题转化为一个多目标优化问题,即利用高效的优化算法优选出一个合理的钢框架-钢板剪力墙结构,之后将这个优选模式通过深度学习的方式生成一个可进行钢框架-钢板剪力墙结构优化设计的BP神经网络.上述优化思路能减少基于分析软件和人工调整的繁复工作,加快和简化钢框架-钢板剪力墙的设计过程.
In order to give full play to the lateral displacement resistance and bearing capacity of steel frame-steel plate shear wall structure,and meet the requirements of the dual lateral resistance system and multi-channel seismic fortification required by the high-rise buildings,it is necessary to design a structure with reasonable failure mode and good energy consumption structure.Most traditional design ideas often require artificial assumptions and modifications.Even the optimization is mostly aimed at a single performance goal or a single direction,which is difficult to obtain a global optimal solution that conforms to the structural multi-performance goal in a relatively short time with the low efficiency of artificial assumptions and modifications.In this paper,based on the simplified calculation model of steel plate shear wall,the plastic design of steel frame-steel plate shear wall is transformed into a multi-objective optimization problem from the perspective of structural optimization design,aiming at the performance of the structure.That is to say,a reasonable steel frame-steel plate shear wall structure is optimized and selected by efficient optimization algorithm,and then a BP neural network which can optimize the design of steel frame-steel plate shear wall structure is generated through deep learning.The above optimization ideas can reduce the complicated work based on analysis software and manual adjustment,and accelerate and simplify the design process of steel frame-steel plate shear wall.
作者
吴星煌
朱南海
陈志强
冯冲冲
WU Xinghuang;ZHU Nanhai;CHEN Zhiqiang;FENG Chongchong(School of Architecture and Surveying&Mapping Engineering,Jiangxi University of Science and Technology,Ganzhou 341000,China;School of Civil Engineering,Xi′an University of Architecture and Technology,Xi′an 710055,China)
出处
《江西理工大学学报》
CAS
2019年第5期22-30,共9页
Journal of Jiangxi University of Science and Technology
基金
国家自然科学基金资助项目(51408276,51768024)
江西省青年科学基金项目(20151BAB216023)
江西省教育厅科技项目(GJJ160618)
关键词
钢框架-钢板剪力墙
结构塑性优化设计
简化计算模型
深度学习
BP神经网络
steel frame-steel plate shear wall
optimal structural plastic design
simplified calculation model
deep learning
BP neural network
