Development of a hysteretic model for steel members under cyclic axial loading. (1st April 2022)

• Record Type: Journal Article
• Authors: Liu, Juncai; Tian, Li; Ma, Ruisheng; Qu, Bing
• Is Part Of: Journal of building engineering. Volume 46(2022)
• Publication Date: 2022-04-01
• Subjects: Hysteretic model -- Hysteretic behavior -- Mathematical expression -- Buckling effect -- Genetic algorithm; Building -- Periodicals; 690.05
• DOI: 10.1016/j.jobe.2021.103798
• Languages: English
• ISSNs: 2352-7102

Abstract

Abstract: Steel members present complex nonlinear behaviors due to the asymmetry of the seismic responses of steel members in tension and compression. This research proposed a hysteretic model for predicting the nonlinear behaviors of hot-rolled steel members under cyclic axial loading. The proposed hysteretic model overcomes the notable limitations of the existing hysteretic model and presents considerable advantages. There are a set of linear segments involved in the proposed hysteretic model and each segment is explicitly defined. The recommended values for the empirical parameters included in the proposed hysteretic model are determined through an optimization analysis based on the genetic algorithm. Compared with the available results of the steel members tested previously by other investigators, the proposed hysteretic model is found to be able to accurately predict the hysteretic curves, peak axial resistances and hysteretic energy dissipations of the steel members. The result comparisons indicate that the proposed hysteretic model realistically presents the complex physical behaviors of members including buckling effects in compression, pinching effects in tension reloading excursion, degradation of post-buckling strength, and degradations of subsequent buckling strength. It is also found that the proposed hysteretic model with the parameters properly selected can successfully replicate the hysteretic responses of the steel members tested previously in spite of their differences in cross-sectional shapes, slenderness ratios and end boundary conditions. Moreover, the genetic algorithm enables to identify the approximate parameters included in the mathematical expressions of the proposed hysteretic model with little computational effort. Highlights: A nonlinear hysteretic model is proposed for high-fidelity prediction. The proposed hysteretic model is validated by experimental results. The proposed hysteretic model is compared against the Marshall model. Optimal parameters of the proposed hysteretic model are recommended.