Probabilistic mechanical modelling of concrete creep based on the boundary element method. Issue 2 (January 2019)

• Record Type: Journal Article
• Authors: Oliveira, Hugo Luiz; Chateauneuf, Alaa; Leonel, Edson Denner
• Is Part Of: Advances in structural engineering. Volume 22:Issue 2(2019)
• Page Start: 337
• Page End: 348
• Publication Date: 2019-01
• Subjects: boundary element method -- concrete structure -- reliability -- response surface method -- viscoelasticity; Structural engineering -- Periodicals; Construction, Technique de la; Structural engineering; Periodicals; 624.1
• DOI: 10.1177/1369433218788088
• Languages: English
• ISSNs: 1369-4332

Abstract

The prediction of the future structural behaviour is an essential activity during the design phase. In this study, a novel numerical framework is proposed for investigating the future structural behaviour of two-dimensional structures. The model utilizes the boundary element method for obtaining the mechanical responses. The constitutive material is admitted to manifest viscoelastic response, which enables it to creep. The input parameters such as material, loads parameters and geometry dimensions are considered to possess random characteristics. A probabilistic criterion is proposed using metamodelling by the response surface method. All these features make the proposed numerical model more realistic. As an application, a specific structure is utilized, which is inspired from the real world. The results demonstrate that small geometric deviations do not necessarily impact the global reliability of the structure. At the same time, load estimations have major influence on the global structural reliability. The numerical framework proposed can be utilized for preliminary investigations on the design phase in order to aid the engineers into the decision-making process. Moreover, these observations demonstrate that the boundary element method can be efficiently coupled to other numerical strategies in order to elaborate new robust numerical frameworks able to represent realistically engineering problems.