Size optimization design of members for shear wall high-rise buildings. (1st December 2022)

• Record Type: Journal Article
• Authors: Lou, Haopeng; Xiao, Zhibin; Wan, Yinyuan; Quan, Guan; Jin, Fengling; Gao, Boqing; Lu, Hongjia
• Is Part Of: Journal of building engineering. Volume 61(2022)
• Publication Date: 2022-12-01
• Subjects: High-rise concrete buildings -- Structural design -- Multiple loading scenarios -- Discrete sizing optimization -- Response surface -- Particle swarm optimization; Building -- Periodicals; 690.05
• DOI: 10.1016/j.jobe.2022.105292
• Languages: English
• ISSNs: 2352-7102

Abstract

Abstract: Structural optimization can lead to efficient designs for high-rise concrete buildings. However, the high computational cost of the structural analysis step hinders its application in practical cases. With regards to this problem, a novel hybrid optimization framework is developed in this study. Discrete variables, code-stipulated constraints and other construction requirements under multiple loading scenarios are considered. The hybrid optimization framework contains two steps: in the first step, a computationally efficient approximation of the optimization formulation is constructed using a combined response surface methodology (RSM). Combining all the design constraints, three sub-optimization problems are proposed based on the relaxation model of the original problem to update the response surface model in the near-optimum region. When the response surface model has sufficient accuracy in the near-optimum region, in the second step, a discrete Particle Swarm Optimization (PSO) technique is applied to search for the optimal solution utilizing the response surface model. By integrating the commercially available ETABS, a dedicated optimization software with an independent interface is developed and details for practical software development are included in this paper. Three examples of high-rise concrete building design are presented to demonstrate the effectiveness of the proposed optimization framework, and the results are compared to those of a general PSO algorithm. It is shown that, compared to the conventional design approach, a 14.3% reduction in material usage is achieved by the hybrid optimization framework. Additionally, the proposed framework significantly improves the computational efficiency compared to PSO. This methodology can therefore be used by engineers to optimize member sizes for high-rise buildings so that stable and sustainable designs can be achieved. Highlights: A discrete sizing optimization framework for high-rise buildings is proposed. A response surface model is constructed and updated to reduce structural analyses. Discrete Particle Swarm Optimization method is used to obtain the final solution. Verification of the new methodology is also conducted.