Multi-performance seismic design of low damage bridge piers.
Thesis DisciplineCivil Engineering
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
Dissipative Controlled Rocking (DCR) is an alternative seismic design strategy to traditional Reinforced Concrete (RC) construction. Instead of the structure being monolithic it consists of separate members clamped together with unbonded posttensioning and replaceable dissipative/damping devices crossing the joints and connecting the adjacent members together. DCR is superior to traditional RC construction in terms of seismic performance because the structure has self-centering characteristics, material damage is limited to the damping devices, and the structure has the ability to have its full capacity restored in a short time frame after a major earthquake by simply replacing the dissipative devices.
An issue with DCR is its low redundancy and robustness in earthquake scenarios such as collapse avoidance (Maximum Credible Earthquake) or multiple earthquake events. This is because the aforementioned characteristics depend on two major components: the dissipative devices and the post-tensioning. Redundancy and robustness of the dissipative devices depends on the number of devices provided, and the devices monotonic and cyclic deformation capacity. The redundancy and robustness of the post-tensioning depends on the number of bars/tendons provided and the remaining deformation capacity of the stressed bars/tendons. The situation of low seismic redundancy and robustness is exacerbated especially when DCR is applied to structures which inherently have low redundancy against seismic loads such as multi-span single-column pier bridges.
This thesis explores ways of modifying DCR single-column bridge piers in order to improve the redundancy and robustness of the system to extreme seismic load scenarios by focussing on passive means to protect the post-tensioning from damage and either lengthening the cyclic life of the dissipative devices provided or providing backup dissipative devices. In particular, the idea of including more sets of dissipative devices and or rocking interfaces and designing the system such that their activation depends on the ground motion intensity and is hierarchical/staged in manner is investigated. The term "multi-performance" is used to describe the piers developed in this research because the pier’s capacity is discretised by hierarchical activation, meaning, that it will have multiple performance levels. The umbrella term to describe all structural types using hierarchical activation and DCR is referred to here as MDCR.
Four ways of implementing hierarchical activation were investigated: having two sets of dissipative devices across one rocking interface, but, where one set is only activated once deformation of the structure has exceeded some specified limit (horizontal hierarchical activation: HHDCR). Having multiple rocking interfaces along the length of the column and dissipative devices across each interface, where, activation of devices up the length of the column depends on ground motion intensity (segmented column DCR: segDCR). Combining a rocking foundation such as a pile-cap with a cantilever DCR column and activating the pile cap rocking once deformation of the structure has exceeded some specified limit to utilise rocking isolation and protect the post-tensioning from overstraining (pile-cap rocking DCR: PCDCR). Finally, the application of hierarchical activation to a complete bridge system where dissipative devices placed outside of the pier are activated after or before activation of the pier dissipative devices.
Three major experiments were conducted as part of this thesis: testing of a 1/3 scale cantilever column utilising DCR and HHDCR; testing of a 2/3 scale cantilever column utilising segDCR, PCDCR and DCR; and testing of a two-span 1/3 scale bridge specimen which used the previously mentioned 1/3 scale low damage column. In addition to these large scale experiments, a rigorous experimental program was also undertaken on the grooved type dissipator. The work in this thesis was not limited to conducting experiments alone. Numerical and analytical work was also conducted in the form of the development of section analysis techniques and 2D multi-spring element modelling techniques for MDCR; undertaking non-linear response history analyses; undertaking parametric analysis; and, developing analytical models and design tools.
The primary outcomes of this thesis are as follows. Achieving hierarchical activation across a single and multiple rocking interfaces. Obtaining experimental evidence of hierarchical activation and the resulting structural behaviour. Comparison of MDCR pier response with conventional DCR pier response. Discovery of unexpected displacement compatibility related phenomena (between the superstructure and substructure and also just within the superstructure) from experiments conducted on a bridge system. Verification of section analysis and multi-spring element modelling techniques for MDCR. Demonstration of the benefits of MDCR compared to DCR through non-linear response history analysis. Determination of the mechanical and geometric parameters which control particular aspects of the cyclic response of MDCR piers. Characterization of the cyclic behaviour (low cycle fatigue and conditions for instability) and geometry of the grooved type dissipator and development of a design procedure for that damping device. Finally, development of a DDBD specific simplified design procedure for MDCR piers to assist practitioners in quick/initial design of SDOF MDCR structures.