Comparison between predictions of the annual cyclic response of a semi-integral abutment using finite and discrete element methods

Resumo

This work compared predictions of the annual cyclic response of a semi-integral abutment retaining a
granular backfill using the Finite Element Method (FEM) and the Discrete Element Method (DEM). Finite element
(FE) and discrete element (DE) models were developed based on the field data collected from an instrumented and
monitored semi-integral abutment. A 15-node triangular FE mesh was used to discretize the backfill in the
simulation with FEM while spherical particles were used to represent the backfill in the simulation with DEM. A
± 5-mm lateral displacement was imposed on the abutment to simulate the effects of expansion and contraction of
the bridge superstructure due to annual temperature variations. The cyclic sequence of imposed lateral
displacements was chosen to simulate the abutment lateral movements after the bridge construction completion in
the summer season. Results showed that a good agreement was found between the numerical simulations using
FEM and DEM. Lateral earth pressures on the abutment and vertical displacement of the backfill surface increased
with annual cycles for both methods. Values of peak wall reaction ratio were similar in both methods while higher
values of settlement were observed using DEM compared to FEM.