Seismic resistance of concrete arch dams including contraction joints based on damage plasticity model

Purpose The proposed model is implemented in a special finite element program for its three-dimensional solid elements (Lotfi, 2009), with its nonlinear dynamic analysis modified according to the HHT-α algorithm, which is fully described. Based on the prepared tools, nonlinear behavior of the Morrow Point dam is studied by applying the invoked damage plasticity concrete model in the presence of contraction joints. For this aim, four cases are considered for the present study. Case A (SF = 1.4), which is based on the linear model, is mainly used for comparative purposes. The other three cases (B, C and D) correspond to the nonlinear model (i.e. damage plasticity model for concrete and nonlinear behavior for contraction joints) with earthquake loading scaling factors of SF = 1.4, 2.0 and 2.6, respectively. Design/methodology/approach In the present study, the application of the damage plasticity model in nonlinear dynamic analysis of Morrow Point concrete arch dam is presented. This model is a single-surface isotropic damage plasticity concrete model based on decomposition of stresses and was proposed in a previous study. It should be emphasized that contraction joints are also considered in the model. The theoretical aspects of the model are initially reviewed, and preliminary verification examples are presented. Thereafter, the HHT-α algorithm is presented for nonlinear dynamic analysis of concrete dams. Findings (1) In relation with displacement response for nonlinear simulations (i.e. cases B, C and D with SF (i.e. Scaling Factor) = 1.4, 2.0 and 2.6), it is noticed that maximum value of response increases as scaling factor becomes larger. This is such that the maximum value reaches an amount −0.35 m (negative refers to upstream direction) for case D. Moreover, the period elongation becomes quite significant. (2) The nonlinear simulation reveals that Morrow Point arch dam can resist a loading scenario based on Taft earthquake excitation even with an scaling factor as high as SF = 2.6. In this case, the maximum value of upstream displacement at dam mid-crest point reaches −0.35 m which is quite high in comparison with usual linear analysis responses. The extension of tensile damages is quite widespread while for compressive damages, it is much less. Moreover, the maximum tensile and compressive stresses for this case are 3.07 MPa and −33.31 MPa, respectively. Originality/value (1) As for the damages plotted at the end of analysis in each case, it is noticed that tensile damage becomes widespread as scaling factor increases. This is such that for case D (SF = 2.6), tensile damage spreads throughout the top portion of the dam body above that initial D/S closed loop occurring for case B (SF = 1.4). This is in both D/S and U/S faces and actually through the whole dam thickness in most parts. In regard to compressive damages, it is observed that it is developed in quite limited portion of dam body for case B. However, it begins to become widespread as scaling factor increases. Of course, most extensive damages corresponds to case D with SF = 2.6 as expected. However, it should be emphasized that even for this case, the amount or extension of compressive damages are much less than corresponding tensile damages. (2) The nonlinear simulation reveals that Morrow Point arch dam can resist a loading scenario based on Taft earthquake excitation even with an scaling factor as high as SF = 2.6. In this case, the maximum value of upstream displacement at dam mid-crest point reaches −0.35 m which is quite high in comparison with usual linear analysis responses. The extension of tensile damages is quite widespread while for compressive damages, it is much less. Moreover, the maximum tensile and compressive stresses for this case are 3.07 MPa and −33.31 MPa, respectively.