Analysis on the shear failure of HSS S690-CWGs via mathematical modelling. (January 2023)

• Record Type: Journal Article
• Authors: Mohamed, Hazem Samih; Elsawah, A.M.; Shao, Yong Bo; Wu, Cheng Song; Bakri, Mudthir
• Is Part Of: Engineering failure analysis. Volume 143:Part A(2023)
• Publication Date: 2023-01
• Subjects: Shear strength -- Bridge girders -- Neural network -- Kriging model -- High strength steel; System failures (Engineering) -- Periodicals; Fracture mechanics -- Periodicals; Reliability (Engineering) -- Periodicals; Pannes -- Périodiques; Rupture, Mécanique de la -- Périodiques; Fiabilité -- Périodiques; Fracture mechanics; Reliability (Engineering); System failures (Engineering); Periodicals; Electronic journals; 620.112
• DOI: 10.1016/j.engfailanal.2022.106881
• Languages: English
• ISSNs: 1350-6307

Abstract

Highlights: The shear strength of HSS-S690 CWGs (τHSS-S690 CWGs ) was investigated. Several mathematical models were employed to predict τHSS-S690 CWGs . A BP-neutral network for predicting shear strength of CWGs was established. Kriging mathematical model shows more accurate results than BP-neutral network. A parametric study on τHSS-S690 CWGs was conducted through 1530 predicted data. Abstract: This paper aims to present a new and accurate design equation by employing the applications of deep learning and mathematical modeling techniques in calculating the shear strength and the failure mode of corrugated web girders (CWG) formed by S690 high-strength steel (HSS). HSS provides slenderer and weight-efficient structures than those would be possible if ordinary-strength steels were used, while the corrugated web (CW) increases the shear stability of steel girders and eliminates the need of transverse stiffeners. Therefore, the CWGs and HSS can be combined into one structure for gaining more structural benefits for civil engineering applications. A finite element (FE) model is firstly established and verified by the available experimental data in the literature. Thereafter, 106 FE models of CWGs are generated and analyzed, then employed in deep learning and mathematical modeling techniques such as the neural network (NN) and Kriging model, respectively. This investigation helps to present a new mathematical model able to predict the shear strength of S690-CWG with almost zero mean square error. The predicted results via the new equation agree well with those via the experimental tests, FE analysis and NN predicted data. Moreover, the comparative results show that the currently proposed equation gives an accurate or even exact shear strength of S690-CWGs than thus via the existing equations in the literature. Finally, an extensive parametric study has been established to investigate the influence of the CWGs' geometrical perimeters on the shear strength of S690-CWGs.