Seismic Hazard Assessment of Four-Column Bent Reinforced Concrete Bridge Pier in Thailand

As part of comprehensive seismic hazard assessment of highway bridges in Thailand, two half-scale full-frames of four-column bent reinforced concrete bridge pier were tested under reversed cyclic static loadings. Four-column bent pier is a typical pier of 5-m to 10-m or short span bridge in Thailand. The first frame represents a typical pier constructed according to Thailand Department of Highways standards. The frame comprises two vertical intermediate columns and two legging (10% inclination) outside columns. The columns are equally spaced along the top cap beam and braced at mid-heights by horizontal beam with relatively large corbels at beam-column connections. The scaled dimensions of a column are square cross section of 0.2 m X 0.2 m and height of 2.50 m. For cap beam, the scaled dimensions are rectangular cross section of 0.2 m X 0.35 m and length of 4.80 m. Reinforcement details are of ordinary types with no specific seismic details. The second frame represents a modified frame in which all corbels are removed. This modification is intended to provide more flexibility to the frame in order to absorb more strain energy by simple and cheap method. Hysteresis loops drawn from the experiments confirm that the modified frame with all corbel removed can store more strain energy than the typical frame with only a slight drop in failure load. The failure mode changes from brittle failure in the typical frame to more satisfactory limited ductile failure (ductility factor of 3.33) in the modified frame. To gain insight into the vulnerability to failure of the typical pier and the modified pier, the variations in concrete compressive strength and reinforcing steel yield strength that might affect the ductility of the structure as well as expected peak ground acceleration (PGA) at a given site must be considered. This task is accomplished by constructing the fragility curve which defines probability of exceeding a set of performance levels as a function of PGA. Two performance levels have been defined- level 1: heavy damage and level 2: collapse. All necessary parameters required to construct fragility curves such as resisting lateral force at yield state, lateral displacement at yield state, and lateral displacement at ultimate state are determined by nonlinear push-over analysis. Fragility curves of the typical pier and the modified pier have been simulated for 8 different sites representing 8 different expected PGAs. It has been found that fragility curves of the two piers have no significant difference. This means that modifying the pier by removing the corbels does not necessary enhance the performance of the pier in resisting seismic forces. In addition, the fragility curves of the two piers in northern Thailand (high seismic hazard area) reveal that these two piers although will not be collapsed, they have 70% chance of being heavily damaged at PGA of 2,500-year return period. Fragility curves are further simulated for four-column bent piers both with and without corbels of varying position of the horizontal braced beam. Identical results have been obtained. For a fixed position of horizontal braced beam, there is no significant difference between pier with corbels and pier without corbels in resisting seismic forces. Four-column bent pier which horizontal braced beam constructed towards the cap beam is the most vulnerable to failure in the high seismic hazard area of Thailand. It has been observed from the simulated fragility curves that this type of pier will certainly be damaged with 25% chance of being heavily damaged and 75% chance of being collapsed at PGA of 2,500-year return period. The results of this study indicate that more sophisticated method rather than removing the corbels is required to enhance the performance of the four-column bent pier in high seismic hazard area of Thailand. Research is currently being conducted at Thailand Department of Highways to find the appropriate strengthening methods