Modeling the fracture progressive process of slopes with a combined finite-discrete element method

In this study, an in-house combined finite-discrete element method (FDEM) software is further developed to investigate the whole progressive failure process of slopes under natural conditions with the influence of discontinuous joints. A fast convergent scheme with local damping and mass scaling is first implemented into FDEM to quickly achieve geostatic stress equilibrium compared with critical damping scheme in conventional FDEM. The strength reduction method prevalent in finite element method together with two developed control criteria are then innovatively implemented into FDEM to not only automatically determine the critical failure surface and corresponding factor of safety but also model the whole progressive slope failure process, including the initiation and propagation of fractures, the formation of failure surface as well as the transport and deposition of blocks and fragments through massively parallel computing based on general-purpose graphic-processing-units. After that, discontinuous joints with various connectivity rates, spacings and dips are implemented into FDEM to investigate the whole progressive landslide process of a natural slope with complex geology. The proposed method is first validated by replicating the slope failure process obtained from a laboratory centrifuge model test. Then, it is applied in simulating the Randa landslide event occurred in the southwestern Swiss Alps, which takes account of the cases with and without discontinuous joints on site and reproduces the whole progressive landslide process. Furthermore, the effects of the connectivity rate, spacing and dip angle of the discontinuous joints on the whole landslide process of the Randa slope are investigated. The associated mechanisms regarding the critical role of the complex joint distributions in the landslide process are also highlighted. The developed FDEM software in this study is expected to provide a powerful tool for predicting and analyzing the complex landslide process under the effects of various crucial factors on site.<p></p>