Optimization Design for Steel Structure Residence Based on the Sequential Quadratic Programming Method

Abstract:

In order to reduce the steel consumption for one of the steel structure residence, a kind of effective approach of optimization design to the original design is given. In this paper, steel consumption of H-section steel beams are taken as objective function, which have relatively larger proportions than others. Design variables are the beam height, web thickness, flange width and flange thickness. The strength, stiffness and stability of the steel beam are taken as constraint condition under the relevant Building Code. By using the sequential quadratic programming method, we set up optimization mathematical model for optimizing the steel structure residence, and then obtain the optimal section size of the steel beams. After the structural optimization design, the steel consumption of GL1 was reduced by 8.24%, compared with those of the original design. Relative to the original design, optimized total steel consumption was reduced by 3.29%. The approach proposed in this paper is an effective method for the optimization design of steel structure residence.

• Periodical: Applied Mechanics and Materials (Volumes 256-259)
• Pages: 680-683
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.256-259.680
• Online since: December 2012
• Authors: Xi Shuang Wu*, Tie Lin Liu, Wen Bo Chen
• Keywords: H-Section Steel Beam, Optimization Design, Steel Consumption, Steel Structure Residence

References

[1] Y.C. Cai: J. Steel Construction Vol. 20 (2005), p.71, in Chinese.

[2] C.G.Yu: D. The Research on Envelop System and Development Policy of Steel Structure Residence In Housing Industrialization. Shanghai. Tongji University (2006), in Chinese.

[3] H.Gong and G.Q.Li: J. Structural Engineers Vol. 20 (2006), p.1, in Chinese.

[4] H.B. Zhang: D. A study on optimization of construction design cost. Zhejiang. Zhejiang University (2011), in Chinese.

[5] J.P.Li: D. A smooth approximation-BFGS method for inequality constrained nonlinear programming.Hohhot. Inner Mongolia University of Technology (2004), in Chinese.