Abstract

Better understanding of collapse resistance of casing and tubing can unlock significant value in support of Asset Life Extension (ALE), support routine Well Integrity assessments in every day work and save significant cost by omitting costly oversized designs. Many operators still use the traditional API collapse model, which were accurate for tubulars produced 50 years ago but now underestimate collapse resistance and predicts typically 80 – 85% of the real collapse pressure. Adding to the excess dimensioning is the standard procedure of applying a safety factor to this prediction.

Early 2000, a joint API/ISO Work Group 2b (WG2b) under the Steering Committee 5 (SC5) for tubular goods reviewed casing and tubing performance property equations. ISO/TR 10400:2007, equivalent to API TR 5C3, presents the results from the extensive testing, and the Klever and Tamano (K&T) model for collapse prediction was found to be most accurate. Building on the test data from WG2b/SC5 group, a model was made for collapse pressure prediction of tubulars - hereafter referred to as the "Ultimate Limit Strength (ULS) model", where the simulation result is a prediction of tubular failure. Its predictive accuracy is calibrated with a complete set of data from 113 actual collapse tests offered by the Drilling Engineering Association (DEA). The ULS model was used to predict collapse strength of 9 5/8 inch 53.5 ppf, P-110 casing, using parameters with probability density functions (PDF) for the relevant type of pipe, e. g., quenched & tempered (Q&T), hot rotary straightened (HRS). The PDFs for each input parameter were obtained by measurements of the 113 samples and compared with the PDFs obtained by the WG2b/SC5 group. Random value generators in a mathematical spreadsheet allowed for Monte Carlo simulations to output 100 000 collapse strength predictions for the 9 5/8 inch casing in question. With confidence level of 97.5%, the basic strength was 9900 psi using PDFs from the DEA data set. Using ensemble PDFs, the basic strength was 9500 psi – 19.5% greater than API's standard rating of 7950 psi.

Performing casing and tubing design, the industry practice is to develop load cases to identify the design limiting loads for the well. Once identified, the pipe selected needs to be investigated for factors reducing the collapse capacity further, e.g. axial / triaxial loads and wall loss from wear and tear. Axial loading is accounted for in the ULS model through the theories of Klever & Tamano. Aspects briefly discussed and not fully incorporated in the prototype ULS model are the linear derating factors considering imposed ovality, casing wear and experimental formulas derived for increased collapse strength of pipe in compression. These were conservatively approximated by polynomial curve fitting of an alternative formulation of yield collapse strength and tried in a version of the prototype model.

You can access this article if you purchase or spend a download.