Experimental evaluation of the static and dynamic characteristics of an offshore crane model

Abstract

The knowledge of the static and dynamic characteristics of cranes is of utmost importance in their design and construction in order to optimise the construction material costs as well as to optimize their response to static and dynamic loadings during operation. Although numerical simulation of the crane response subjected to various loading conditions are routinely undertaken in the design stage, experimental verification of these numerical results are rarely performed due to the high cost involved. Experimental verification of the static and dynamic characteristics of these cranes can however be performed at a significantly lower cost on their scaled-down models. The results from the scaled-down models provide useful insights into the static and dynamic performance of these cranes. In the work presented herein, a scaled-down model of a boom angle luffing crane prototype, typically employed in offshore engineering applications, was developed using dimensionless p parameters. The model, which has geometric and dynamic similarities with the prototype, was numerically and experimentally examined for its static and dynamic characteristics. The numerical analysis of the model was undertaken using a commercially available finite-element computer program, ANSYS. This program was utilized to compute the stiffness and stresses in the model subjected to static loads, as well as the natural frequencies and mode shapes, which represent the dynamic characteristics of the model. An experimental rig was fabricated based on the scaled-down model, and measurements were performed to verify the computational results obtained from the finite-element analysis. The static strains measured at various positions on the boom were found to be within 14% of those obtained numerically. The natural frequencies and mode shapes of the model crane were obtained using modal testing technique and were found to be in good agreement to those obtained numerically, with discrepancies within 3% for the first three modes.