Multi-objective optimization design of coupled wall structure with hybrid coupling beams using hybrid machine learning algorithms

Abstract

A new hybrid coupling beam incorporating buckling restrained energy dissipaters and viscous dampers was previously recommended. The beam can effectively reduce the floor acceleration and interstory drift ratio of coupled wall structure, and has great application prospects. However, the optimal design of a structure with hybrid coupling beams needs to consider multiple objectives, and these objectives often conflict with each other. At the same time, multiple parameters need to be optimized, resulting in a corresponding increase in the time-cost. This usually brings challenges to structural design. To address this issue, this study presents a hybrid intelligent optimization framework based on multi-objective particle swarm optimization (MOPSO) and machine learning techniques. The developed framework uses machine learning technology to quickly predict the seismic performance of structures with hybrid coupling beams, and obtain a design solution set through multi-objective particle swarm optimization. The proposed framework is demonstrated by considering a case study of a 12-story coupled wall structure with hybrid coupling beams, and it is shown that compared with the initial stage of optimization, the seismic performance of optimized structure is simultaneously improved at close cost. The floor acceleration and interstory drift ratio of the optimized structure are reduced by 11.5% and 38.7%, respectively. Finally, a formula is proposed to quantify the optimization result of pareto solution, so as to make an optimal structural design scheme. In practice, the proposed framework can provide guidance for realizing rapid and accurate about multi-objective optimization of the coupled wall structure with hybrid coupling beams.