Shake table experiments for the determination of the seismic response of jumbo container cranes

Container cranes represent one of the most critical components of ports worldwide. Despite their importance to port operations, the seismic behavior of cranes has been largely ignored. Historic data show that the destruction of a port during an earthquake can have a significant impact on business interruption losses, such as at the Port of Kobe in 1995, as well as complicate rescue efforts in affected regions, such as in Port-au-Prince, Haiti in 2010. Since the 1960s, industry experts have recommended allowing cranes to uplift, believing that it would limit the amount of seismic loading. However, modern cranes have become larger and more stable, and the industry experts are now questioning the seismic performance of modern jumbo cranes.;The main goal of this research is to experimentally investigate the seismic behavior of container cranes from the elastic behavior through the inelastic behavior utilizing the 6 degree-of-freedom shake tables at the University at Buffalo. Due to the complexities of modeling uplift, the characterization of uplift and derailment behavior is of particular interest. Additionally, no experimental studies have ever been conducted on container cranes to gather information about buckling, yield and collapse. This research project addresses these issues.;To characterize the seismic behavior of container cranes, the testing was divided into two phases. The first phase of testing was conducted on a 1/20th scale model and focused on the uplift and elastic behavior. The data collected, confirmed that a simple tipping analysis is sufficient to predict when derailment will occur. The results from the Phase I test also indicate that torsion has little effect on the overall response of a jumbo container crane, suggesting that 2D finite element models are sufficient for analysis. Additional results suggested that the top structure has little influence on the critical response quantities, indicating that simplifications could be made to the structure while preserving the dynamic characteristics.;In support of the experiments, finite element models were created to determine what simplifications could be made to the structure to aid in testing. The results of the finite element models indicate that it is possible to remove the top structure and still have a dynamically equivalent model. This result is advantageous to the port community because it suggests that simplified models can be used for analysis and qualification tests, which can save time and money.;The Phase II test was designed to be representative of a modern jumbo crane. It was also designed such that no inelastic action would develop prior to uplift (as is the common design practice). During testing the crane yielded, buckled and reached an unstable state after uplift, which challenges the conventional wisdom that uplift will provide a limit to the seismic forces. Two different boundary conditions were used: pinned and free to uplift, for the validation of finite element models with different boundary conditions. There were cases for which the pinned boundary condition led to higher forces, suggesting that cranes in earthquake zones should not be pinned except when the weather dictates otherwise. The test specimen was subjected to a suite of ground motions with various combinations of components. The data show that there are cases for which the inclusion of a vertical component will result in the most extreme load case. This result suggests that it would be prudent to run models with and without a vertical component to find the worst case. All of the test and analysis results are used as the basis for recommendations to the port community.