Early-age to long-term numerical simulation of concrete members tested in adjustable restraining frames

Abstract

Avoiding or limiting crack widths in massive concrete elements can be a challenging problem owing to the physical phenomena inherent to hardening processes occurring at early ages. In restrained elements, the increase of temperature at an early stage, due to the heat of hydration of cement, and the consequent cooldown at a stiffer stage, results in tensile stresses, which may lead to cracking or reduce the tensile capacity to sustain additional loading in the long-term. A recurrent set of questions, made by designers and scientists, is: do the concrete stresses due to hydration heat relax in the long-term? or do they need to be superposed to the stresses owing to long-term effects and applied loading? and, do the hydration heat effects influence the long-term crack widths? An extensive experimental campaign was recently conducted at TU Graz, to help answer these questions. But a different set of questions persists: how accurate are finite-element analysis methodologies (simulating the viscoelastic, time-dependent, concrete behaviour, cracking and bond-slip effects, at the macro-scale), to predict the developed concrete stresses and crack widths, since the early ages until the long term? The present papers intents to contribute to answering the latter, by performing the finite-element analysis of the beforementioned experimental campaign made at TU Graz, and by discussing the comparison between experiments and numerical analyses.