Discrete versus Smeared versus Element-Embedded Crack Models on Ring Problem
Abstract
Three different concepts in finite-element modeling are compared by analyzing a thick-walled concrete ring subjected to an internal pressure. These are the discrete crack, the fixed smeared crack, and an element-embedded crack model called the inner softening band (ISB) model. The results are compared with experimental studies on rings of both normal- and high-strength concrete mixtures with and without spiral reinforcement. The ISB analysis gives the best results with respect to explaining the fracture mechanism as well as consistency in prediction of the ring capacity. The smeared crack model also provides satisfactory results, but in the light of robustness in geometrical modeling and requirement of minimum amount of input data (which all have a physical relevance) the ISB concept in this case is more reliable. The study exposed some deficiencies of the discrete crack method based on the fact that the localization processes and the strain redistribution in the body are often unknown. This also underlines caution in assuming symmetry when softening occurs on a structural level.