General Reliability Analysis of a Tension Leg Platform

Summary

The following objectives have been identified for this project:

1. The TLP is compliant in the surge, sway and yaw motions. That is, the TLP moves with the waves in these degrees of freedom rather than resisting the wave action. The resulting horizontal displacement (in the surge motion) can be a substantial fraction of the wavelength and needs to be considered in the calculation of the induced wave forces. That means that the TLP analysis should be done in the displaced position. The effects of evaluation the maximum tendon group forces when the multiples of the wave frequency coincide with the heave and pitch natural frequencies.
   For this purpose, a deterministic parametric analysis (a study of the variation of the maximum tendon group forces with selected parameters) will be conducted for the wave height and period, the inertia and drag coefficient and the modulus of elasticity of tendon steel. These parameters have been selected for the parametric studies because there are uncertainties associated with them and ay nee to be randomized for the reliability analysis.
2. There is a need for improved knowledge of the extreme values of the forces in the tendons and their probability distribution to design the tendons. This knowledge can only be achieved through reliability (probabilistic) methods since there are only six TLPs that have been designed and/or installed to date. Thus, the statistics (cumulative distribution function) of the maximum tendon group force under the action of regular and irregular waves using advanced mean value first order second moment (AMV-FOSM) and first order reliability method (FORM) will be explored.
3. The Airy (linear) wave theory is valid up to the mean water level and different approximations for calculation of wave kinematics up to the wave profile were investigated by Mekha (1994). Chakrabarti’s non-linear stretching approximation will be used in this work to study the reliability implications of the calculation the wave kinematics up to the free water surface rather than to the mean water line.
4. The foundation can reduce the net stiffness of the tendon because it is not rigid. There is a need to investigate the foundation effects on the tendon stiffness. This will be achieved as a first approximation by reducing the axial stiffness of the tendon steel (10 – 50%). The reduction of the axial stiffness can also result from increasing tendon length (i.e., in deeper sea than the 5000 ft considered in this work).
5. The reliabilities and probabilities of failures given by mean value first order second moment (MV-FOSM), AMV-FOSM, FORM and second order reliability method (SORM) will be compared to evaluate the variability of the results.

Related Publications:Acquaah, C.G. and Gilbert, R.B., “Probabilistic Analysis of Tendon Loads for a TLP in Deep Water,” Proceedings of Advances in Probabilistic Mechanics and Structural Reliability, ASCE, Worchester, pp. 162-166, 1996.