Influence on uncertainty of earthquake response analysis results by initial particle arrangements and cohesion parameters in extended distinct element method
Document typeConference report
Rights accessOpen Access
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder
Following the occurrence of extremely large earthquakes, such as the Great East Japan Earthquake, the level of design for earthquake ground motion in nuclear power plants has been enhanced. Additionally, the quantitative evaluation of the seismic performance of critical facilities, such as nuclear power plants, and earthquake-induced failure of surrounding slopes are becoming increasingly important as deterministic approaches in regulation. However, evaluation of other aspects besides the design for earthquake ground motion in probabilistic risk assessment (PRA) needs to be conducted voluntarily by the corporation. For the earthquake response analysis, including the seamless transition of the slope from continuum to dis-continuum, the extended distinct element method (EDEM) is an effective approach; however, EDEM is characterised by initial particle arrangement uncertainty. Therefore, we investigated the uncertainty in the EDEM results with respect to failure timing and region. Although essential in the evaluation of impact force in the PRA framework, there are few researches regarding the uncertainty of impact force on the wall of the reactor building after slope failure caused by numerous initial particle arrangements. Furthermore, reducing the computational time is crucial in PRA. Hence, the parameters that do not have an influence on the EDEM results can be omitted, resulting in their dispersion and a reduction in the computational time. This research aims to investigate the impact force uncertainty caused by initial particle arrangements and the influence of cohesion uncertainty. For the former, we conducted 50 numerical simulations for the uncertainty of EDEM results caused by the initial particle arrangements. For the latter, we conducted 50 numerical simulations with two uncertainty factors, namely, cohesion and initial particle arrangement. The simulation results revealed that the largest and second largest loads on the wall occurred in two cases, namely, when there were single particles impacting the wall and when there were group particles impacting the wall. Additionally, the uncertainty caused by cohesion was less than that arrangement when the coefficient of variation was 0.1. Thus, the cohesion uncertainty can be ignored if it is somewhat small.