摘要
实际工程中,门式刚架设计大多是依照设计人员经验进行试选截面,再进行强度、刚度、稳定性等一系列验算完成设计,一般并不进行截面比选和优化,往往会造成钢材大量浪费.文中设计一种基于混合编码和个体特殊属性保护策略的改进遗传算法,对门式刚架截面进行优化.通过实际算例分析,相较传统设计方法,改进遗传算法可以优化接近5%的门式刚架总重量,从而为门式刚架的设计提供一种新的可行优化方法.
The traditional portal frames structure design is based on the experience of designers to try a fit cross-section, and then a series of strength verification and others to complete it. In most of time, due to lack of experience with designers, the design could waste lots of steel. In this thesis, we design an improved genetic algorithm based on hybrid coding and individual special attribute protection strategy, which could optimize the cross-section of steel portal frames. According to the analysis of actual calculation, an improved genetic algorithm is able to optimize the total amounts of 5 % of steel portal frames, compared to the traditional design. Therefore, an improved genetic algorithm provides a new feasible optimization for the design of the steel portal frames.
作者
李彬
汪小平
张柳春
朱南海
LI Bin;WANG Xiaoping;ZHANG Liuchun;ZHU Nanhai(School of Architecture and Surveying ＆ Mapping Engineering,Jiangxi University of Science and Technology,Ganzhou 341000,China)
出处
《江西理工大学学报》
CAS
2018年第5期9-14,共6页
Journal of Jiangxi University of Science and Technology
基金
国家自然科学基金资助项目(51408276)
江西省青年科学基金资助项目(20151BAB216023)
江西省教育厅科技项目(GJJ160618)
关键词
门式刚架
改进遗传算法
混合编码
截面优化
portal frames
improved genetic algorithm
hybrid coding
cross-section optimization
