A laboratory reduced scale model of an offshore jacket structure is described. The structure was excited with a miniature vibration exciter with random excitation source input. Multiple accelerometers were mounted along the main columns of the test structure, and acceleration response signal acquired through a multichannel data acquisition system. The natural eigenfrequency and eigenvector were determined from the ARMAV (Vector Auto Regressive Moving Average) model, and the modes identified from the response signal using the subspace state space system estimation method. Resultant natural eigenfrequencies were plotted on a frequency stability diagram. The structural resonance mode was identified from the frequency stability diagram, and the corresponding eigenvectors and natural eigenfrequencies quantified for average and standard deviations.
The size and complexity of structural designs of structures for operation in the offshore environment have increased considerably over the years. Increased construction and installation costs had lead to a design requirement of minimum weight of structure used. These implied slender offshore structural designs and are consequently more flexible when subjected to sea waves loading. While operating in a continuous, hostile and corrosive environment, the potential degradation of structural components and foundation degradation are of concern to the structures' owners and operators. These have economical and safety implications. Additional damage can also occur due to loads induced during vessel docking operation. This paper examines the identification of potential damage to structural element of a typical 4 leg offshore jacket structure. The structure considered in this work is typically installed and are in current use in the South China Sea. One method of structural verification conventionally used is by visual inspection. This method however be impractical due to its high cost, and is labor intensive. It is also almost impractical for inspection at deeper parts of the structure.
