Session: 04-03-01 Rigid Risers

Submission Number: 180247

Small-Scale Tests on the Dynamic Compression of Lazy-Wave Risers: Experimental Results 

Dynamic compression in risers is one of the relevant phenomena to be considered in risers structural safety. Caused by motions imposed by the FPSO driven by surface waves, negative effective tensions may appear along the riser. If these negative effective tensions surpass a certain critical level, a local quasi-static instability (i.e. buckling) may take place, the riser being suddenly and repeatedly attracted to new geometric configurations. Post-critical behavior, possibly establishing high local curvatures, can then lead to structural damage. Consequently, theoretical predictions for the critical buckling loads are desirable from the design point of view. While analytical formulas for both in-plane and out-of-plane buckling have been derived and experimentally validated for risers in free hanging catenary, their validity for lazy-wave risers (LWRs) has not yet been studied.
 
This paper addresses an experimental study aiming at: (i) assessing the validity of the existing analytical formulas to predict the critical buckling load for LWRs; (ii) characterizing the relevance of the dynamic compression phenomenon in the presence of other ones, such as the so-called Heave-Induced Vortex-Induced vibrations (HIVIV); (iii) building an experimental database to be used as benchmark for numerical codes, aiming at assessing their capability to predict the dynamics of a LWR in such complex scenarios. A small-scale LWR model was designed, composed of an 8-meter-long silicone pipe filled with stainless steel microspheres to which five 3D printed floaters were attached. The displacements of the LWR were directly measured using an underwater optical tracking system, also capturing the 6-DOF motions of the rigid floaters.

The LWR static configuration was designed considering the dimensions of the test facility. The critical loads for both in-plane and out-of-plane buckling along the LWR were determined using the existing analytical closed-form formulae. Such formulae give the critical load as function of the frequency and static curvature. The first, by Aranha et al. (2001), considers the instability phenomenon restricted to the original LWR plane. The second one, by Ramos & Pesce (2003), relaxes such restrictions, allowing the structure to twist and buckle outside that plane. The dynamic response of the LWR was investigated using OrcaFlex™, and, by comparing the resultant effective tension envelopes to the predicted critical loads, stability thresholds were obtained as function of frequency.

The text matrix was defined based on the theoretical stability threshold, associating it with the imposed top motion amplitudes, hereby referred to as "critical" ones. In total, 13 frequencies, at 4 different amplitudes, were imposed at the top of the LWR, leading to 52 different experimental scenarios. The experimental observations showed unequivocal agreement between the predicted critical amplitudes of imposed harmonic motion and the conditions leading to buckling in the LWR model. 

For sub-critical amplitudes, the LWR model exhibited smooth motions, dominated by a few natural modes, whose amplitudes were seen to increase as the imposed amplitudes neared the predicted critical values. For post-critical amplitudes, still close to the theoretical threshold, weak instability was noticed, and out-of-plane buckled configurations with a unimodal character emerged. In these scenarios, clear stationary nodes appeared in the buckled configurations. Different than when considering sub-critical amplitudes, however, the motions were no longer smooth and the top portion of the LWR displayed high local curvatures, in a repeated and sudden manner, with the frequency of the imposed motion. By further increasing the magnitude of the imposed post-critical amplitudes, vigorous instabilities occurred in the top portion of the LWR. At each frequency of the imposed motion, a different, high-amplitude and multimodal buckled configuration occurred repeatedly, and clear nodes were no longer established. For both sub- and post-critical amplitudes, HIVIV were always observed, playing an important role in the LWR dynamics, particularly in the sag region. The paper will contain a formal analysis of the experimental results focusing on the instabilities caused by the dynamic compression phenomenon, from which design criteria may be drawn.

Presenting Author: Vitor Schwenck Franco Maciel University of São Paulo

Presenting Author Biography: Post-Doctoral researcher at Escola Politécnica of the University of São Paulo. Currently conducting research on pipes conveying fluid, passive suppression of vibrations using nonlinear energy sinks and other topics in Offshore Engineering such as riser dynamics and stability.

Authors:

Vitor Schwenck Franco Maciel University of São Paulo
Celso Pupo Pesce University of São Paulo
Pedro Cardozo De Mello University of São Paulo
Guilherme Rosa Franzini University of São Paulo
Victor Matias De Biaggi Tucci University of São Paulo
Letícia Siqueira Madi University of São Paulo
Cristiano Emilio Justiniano University of São Paulo
Lívia Rampinelli Bozzo University of São Paulo
Raquel Almeida Siqueira Santos University of São Paulo
Rodrigo Provasi University of São Paulo
Guttorm Grytøyr Equinor Group
Rodrigo Do Nascimento Carvalhal Equinor Group