GA-Based Optimization of Steel Moment Frames: A Case Study

HALL, BRIAN S., GA-Based Optimization of Steel Moment Frames: A Case Study (Under the direction of Dr. Abhinav Gupta). Moment resisting frames are utilized in steel construction in order to provide overall stability and to withstand lateral forces related to wind and earthquake loadings. While the industry focuses on producing more economical moment frames by optimizing for the weight of steel, it should be noted that because of present trends the labor cost associated with rigid moment connection fabrication governs the total cost of these frames. Therefore, the objective of this case study is to discover alternative ways of producing more cost-effective moment frames without compromising overall stability. This is achieved by considering both the cost of steel and the cost of connections within the design process. In order to reduce cost, rigid connections within the frame are replaced with standard pinned connections, and member sizes are increased where needed, a method unlike the current least-weight design approach. Optimization techniques are developed in order to identify the most advantageous locations of the remaining moment connections. A Genetic Algorithm, interfaced with a Java-based frame analysis program, is utilized to produce these optimal solutions. At the present time, the consideration of connection-related costs is crucial in determining the most cost effective moment frames. By developing an optimization technique that considers both the number of moment connections and the total weight of the frame, the least weight optimal solution can be improved upon drastically, with an almost 50% reduction in the total cost and an approximately 60% increase in total weight. The unorthodox connection arrangements produced are important, as they provide alternative designs that, while functional, are not currently explored by those in practice. GA-Based Optimization of Steel Moment Frames: A Case Study

基于遗传算法的钢框架优化：一个案例研究（在阿比纳夫·古普塔博士的指导下） 抗弯框架用于钢结构建筑中，以提供整体稳定性并承受与风和地震荷载相关的侧向力。虽然行业通过对钢材重量进行优化来致力于生产更经济的抗弯框架，但应该注意的是，由于当前的趋势，与刚性抗弯节点制造相关的劳动力成本决定了这些框架的总成本。因此，本案例研究的目的是在不影响整体稳定性的前提下，探索生产更具成本效益的抗弯框架的替代方法。这是通过在设计过程中同时考虑钢材成本和连接成本来实现的。为了降低成本，框架内的刚性连接被标准的铰接连接所取代，并且在需要的地方增加构件尺寸，这是一种不同于当前最小重量设计方法的方法。开发了优化技术以确定剩余抗弯节点最有利的位置。一种与基于Java的框架分析程序相结合的遗传算法被用于生成这些最优解。目前，在确定最具成本效益的抗弯框架时，考虑与连接相关的成本至关重要。通过开发一种同时考虑抗弯节点数量和框架总重量的优化技术，最小重量最优解可以得到极大的改进，总成本降低近50%，总重量增加约60%。所产生的非传统连接布置很重要，因为它们提供了一些替代设计，这些设计虽然可行，但目前在实践中尚未被探索。基于遗传算法的钢框架优化：一个案例研究