Dynamic monitoring of an offshore jacket platform based on RTK-GNSS measurement by CF-CEEMDAN method

Offshore platforms are constantly exposed to dynamic forces by winds, waves, currents and earthquakes, which can easily lead to damage to the platforms during their serviceable lifetime. The real-time and continuous dynamic monitoring of offshore platforms in service plays an important role in reducing the risk of structural failure, instability and even destruction. In this study, a two-step approach was presented to capture the dynamic response of an offshore jacket platform based on the real-time kinematic global navigation satellite system (RTKGNSS), which combines a Chebyshev filter and complementary ensemble empirical mode decomposition with adaptive noise (hereinafter referred to as CF-CEEMDAN). In the first step, the RTK-GNSS technique was employed to measure the dynamic displacements of the offshore platform under ambient excitations. Different from traditional accelerators, this technique has no unavoidable errors caused by double integration and directly measures the real-time and continuous dynamic displacements of the offshore platform at site. Next, a CF-CEEMDAN filter was developed to weaken the influence of the background noise in the RTK-GNSS signals. In the second step, the natural excitation technique and eigensystem realization algorithm (NExT-ERA) was applied to estimate the modal parameter (i.e. the first natural frequency) of the platform with filtered RTK-GNSS signals. In order to evaluate the efficiency of the proposed approach, a nonlinear signal with additive noise was firstly introduced to assess the effect of the CF-CEEMDAN method, and a finite element model of the platform was established to predict the natural frequency for comparison. The results prove that the CF-CEEMDAN filter not only has a better noise reduction effect than single CF or CEEMDAN method but also effectively improve the accuracy of RTK-GNSS technique to monitor the dynamic displacements of offshore platforms. Furthermore, the natural frequency obtained by experiment and NExT-ERA analysis is consistent with the predicted value based on finite element model.

Automated summary

This study proposed a two-step approach using RTK-GNSS technology to monitor the dynamic response of offshore platforms in real-time, with the CF-CEEMDAN filter to reduce background noise interference. Experimental results demonstrate that this method effectively improves monitoring accuracy and reduces noise interference.