Computational Study of Granular Debris Flow Impact on Rigid Barriers and Baffles

Debris flow is one type of natural terrain hazards that is particularly destructive. In order to protect the public, mitigation measures such as rigid barriers and baffles are deployed which intercept debris flows. The current understanding of interaction mechanisms between debris flow and protection structures is limited, and results in designs that are not cost effective. In particular, practitioners are unable to grasp the impedance resulting from debris flow baffles since their impedance effect is not known. This study aims at improving the understanding of interaction mechanisms between debris flows and different types of rigid protection structures, and to quantify the effect of baffle configurations on the reduction of peak impact force to downstream structure.;Numerical analysis using the Discrete Element Modelling (DEM) is adopted to achieve the research objectives. DEM is used to carry out a series of impact scenarios which consist of different configurations of barriers and baffles. Input parameters are calibrated by physical tests and numerical experiments.;Based on the DEM study, five key interaction mechanisms are identified, namely frontal impact, runup, pileup, slit flow and overflow. The peak impact pressure is found to be caused by frontal impact which has relatively short duration. The runup process is accompanied by the observed secondary peak impact pressure and peak overflow discharge. In contrast, the overflow is driven mainly by gravity at pileup stage, and the discharge rate is slower. The pileup process dissipates most of the momentum of impacting landslide debris such that the barrier does not experience the dynamic impact load. The pileup process could only occur at transverse blockage not less than 50%. The discharge rate by slit flow is considerably higher than overflow based on realistic geometry of slit dams and baffles.;An optimum spacing between baffle rows is identified to minimize impact on downstream barrier by frontal impact and overflow, thus achieving minimum peak impact force. The baffle height needs to be higher than the frontal flow depth to effectively decelerate the frontal material. For practical considerations, transverse blockage is recommended to be low (i.e. not more than 30%). A reduction of barrier stiffness does not have significant effect on the peak impact pressure. Based on the above findings, a set of recommendations regarding the configuration of baffles to achieve optimum reduction of impact force is proposed.