Damage Identification And Fatigue Life Estimation For Jacket Platform

Early researches are mainly focused on monitoring the stress and acceleration at some important point, analyzing the fatigue life and making assessment of the integrity. As the development of the theories and techniques of parameter recognition, damage analysis, safety assessment and the emergence of high-resolution sensor and instrument, computer hardware and software, the crack detection and real-time monitoring of offshore structure develop quickly. A crack in the structure will change the structural parameters. And the dynamical characteristics can sensitively reflect the shift of the structural parameters. Thus, the dynamic detection is an effective way to find defects.(1) A model test is carried out. The frequency, FRF and acceleration history for different damage is recorded. Natural frequency can be easily measured, but it can only detect breakage of beams in the structure. Mode shape can provide more information of the crack. It can be used to predict the location and the magnitude of the crack. But it needs a large number of survey points to obtain the modal shapes. FRF-data can be used to detect any breakage of beam in the offshore structure.(2) The wavelet transform has an ability to present the local characterics of signal in both frequency and time domain. It can zoom in and recognize the signal abrupt changes of structural responses under different scales. An analysis method of structural damage identification is presented based on .the wavelet transform for the structural vibration displacement mode difference between the intact structure and the damaged structure. The abrupt change of wavelet coefficients indicates that there is some damage occures. The present method is validated through a cantilever beam and a offshore platform.(3) The damage at different locations can make the parameters change differently. However early damage (usually very small) can only slightly change such parameters. Wavelet transform can be employed to detect the subtle shift of the parameter. The present paper aims to use the Wavelet transform for the FRF and acceleration response history to detect the small damage. The model test is carried to verified the proposed method.(4) Daubechies-6 wavelet is used as the mother wavelet. And 6-layer multi-resolution analysis are proceeded. When the crack ratio increases, the singularity of the wavelet coefficients becomes more distinct. If different wavelets are chosen as the wavelet mother function, the wavelet transform results will be different. The Daubechies-6 wavelet can efficiently used to find the singularity of the wavelet coefficients in the multi-resolution wavelet transform analysis.(5) Wavelet packet transform can be used to find the singularity of the signal, such as FRF data and acceleration response, so as to predict early cracks in the structure. An optimized mother wavelet is chosen for the early crack detection in the offshore structure using FRF data or acceleration signal. A crack ofα=0.1 in the bottom of the structure can be detected by wavelet packet transform of the FRF or the acceleration signal(6) Finally, some future suggestion is lined.