Virtual sensing of offshore tubular structures using Kalman Filter estimation

This paper presents a continuation of previous work on digital twin approaches for structural integrity management. Among the various components of a digital twin, virtual sensing represents an indispensable functionality. A widely adopted method for virtual sensing is the modal expansion approach, which assumes that the responses of a dynamically loaded structure can be reconstructed by a finite number of mode shapes with associated amplitudes. Using known mode shapes, the dynamic responses at non-instrumented locations are inferred by estimating modal amplitudes based on physically sensing data from instrumented areas. Conventionally, this estimation requires the number of instrumented locations to be greater than or equal to the number of mode shapes used to approximate the structural dynamics. To address this limitation, the present study explores the use of a Kalman filtering method to estimate modal amplitudes, examining its ability to work with fewer instrumented locations than mode shapes. Experimental data collected from lab-scale tubular structures are employed to demonstrate and verify the feasibility of the proposed approach.