Study On Modal Parameter Identification Of Offshore Platform Structures Based On Stochastic Subspace Algorithm

To avoid major malignant accident, it is necessary to conduct periodic/aperiodic detection and safety assessment for offshore platforms with the action of various environmental forces, complex structure and high cost.So effective health detection for offshore platform, particularly those in old age or in overterm service, is more significant. The structural health monitoring technology based on the response data of the vibration test which is a global detection method is able to construct a comprehensive testing for offshore platforms, of which modal parameter identification is fundamental and critical. Therefore, the accurate modal parameter for offshore platform is particularly important.Early methods of modal identification were developed for the frequency domain. Because of limitations in the frequency resolution of spectral estimates and leakage errors in the estimates for the frequency domain methods, the new trend is to employ either input-output or output-only time-domain modal identification methods, of which the PRONY, Eigensystem Realization Algorithm(ERA) and SSI are more successful methods in other areas.With the aim at develop a effective identification technique for modal parameters of offshore platforms more exactly, the thesis studies on how to identify modal parameters more efficiently and improves existing modal parameters identification technologies by investigating the difference between SSI/data and SSI/cov and relation between Hankel matrix and noise. The main contributions are as follows:1. Based on the data associated with different excitations (impact loading, step relaxation, white noise loading), Prony, ERA and SSI/data are compared. It is proved that SSI/data has a better capacity of identifying the modal parameters of structures under different excitations (impact loading, step relaxation, white noise loading).2. The data-driven and covariance-driven stochastic subspace identification traditionally is thought to be consistent with each other theoretically and practically for modal identification. The paper investigates the reason that probably produce the difference,then confirms the reason of difference, the numerical study demonstrates that data-driven stochastic subspace identification method outperforms the covariance driven subspace identification method not only on accuracy of identification parameter but also on capacity of identifying weaker mode.3. The paper deduces theoretically the relation formula between noise and Hankel matrix of data-driven subspace identification method. And the paper also proposes a verification procedure to justify the noise can be eliminated properly by data-driven subspace identification method with selected Hankel matrix, the procedure includes SVD, stability diagram and finite element result (FE). Finally based on data associated with numerical study and jacket-type platform vibration test separately, we demonstrate systematically that data-driven stochastic subspace identification method with non-square Hankel matrix has better capacity of denoising and estimating the modal parameters with higher accuracy.4. With the data collected from a jacket-type physical model under different excitations(including impact loading, step-relaxation loading, white noise loading) and test data of realistic offshore jacket-type platform in Bohai sea in China, this paper demonstrates that SSI/data can be applied to modal identification of offshore platforms and be able to finish it better, and different strong modes can be excited by different excitation.