Reliability Analysis of Deep-Water Floating Structures

Allin Cornell (associated P.I.), Satya Kumar (graduate student)

1. Joint metocean criteria for deep-water sites,
2. Evaluation and propagation of uncertainty in structural and hydrodynamic properties (loads, damping, etc.),
3. Comparison of analytical models with wave tank data,
4. Parametric reliability analysis across different response quantities, structural concepts, and sites, and
5. Calibration of probability-based design codes for floating structures.

 

1. Efficient prediction of response extremes and fatigue; can include nonlinear diffraction loads from M.H. Kim program and/or WAMIT, both at OTRC,
2. Base case uses OTRC theme structure,
3. Analytical models calibrated from wave tank results,
4. Decoupled metocean criteria can be used by other investigators for different responses, other concepts,
5. Contours of critical wave parameters useful to suggest experimental design of wave tank studies.

 

1. Metocean Environmental Statistics: Joint wind-wave-current probability distributions have been estimated for a generic Gulf of Mexico site. These parallel models used in a recent JIP on TLP code calibration,
2. 100-Year Joint Contours: With newly developed concepts from first-order reliability methods (FORM), 100-year wind-wave-current contours have been found. Based on FORM these contours will contain the 100-year response, regardless of structural model or response quantity,
3. Stochastic Analysis of Extreme Response: Non-Gaussian random process models are developed to estimate extreme and fatigue statistics analytically, avoiding need for costly response simulation,
4. Reliability Analysis and Code Calibration: Results of FORM studies are used to calibrate deterministic design rules for floating structures.

Significant research in floating structure reliability is carried out by the oil industry itself, risk management consultants (e.g., Risk Eng. Inc.) and ship classification agencies (e.g., ABS, DNV, Bureau Veritas, Germanicher Lloyd.) Related university work in both basic hydrodynamics (e.g., MIT, U. Michigan) and reliability applications (e.g., UC Berkeley, Norwegian Inst. Tech., Stanford).

Joint contours decouples environment and response; convenient basis for joint criteria in codified design (e.g., API-2T floating structure recommendations). Efficient prediction of response statistics aids (a) understanding of complex nonlinear hydrodynamic models; (b) comparison with wave tank data; (c) study across different concepts, sites.

October 1993 to September 1997

Most of the project efforts have been directed towards the OTRC researchers and their students providing:

1. Reliability software SHASYS for FORM analysis,
2. Limit state routines, modeling extremes and fatigue, for use in FORM codes(SHASYS, PROBAN),
3. The routine HERTRA to model nonlinear vibrations, applied at OTRC to model collisions between pairs of flexible cylinders,
4. A routine with metocean probability model for Gulf of Mexico site, also suited for use in FORM,
5. Post-processor TFPOP, combining 2nd-order hydrodynamic loads with user-defined 6DOF structural model,
6. The algorithm IFORM, which produces environmental contours for experimental design and prediction of floating structure response.

Project results included at Hydrodynamics Workshop (April 1994), Stanford Affiliates Meeting (May 1994). In-house Industry tutorials, research overviews: Statoil, Saga, Norks Hydro (Summer 1994, 1995 in Norway), Mobil (Jan. 1995), Chevron (March 1995), Exxon (June 1995). Research results presented in scholarly publications and reports.