ABSTRACT
The quantification of risk using reliability based procedures offers significant safety-related and cost benefits. In order to progress this methodology for design of wells, accurate capacity prediction models for OCTG and associated statistical data for the model parameters are required. In this paper, we focus (a) on the development of a predictive model for burst failure of OCTG with naturally occurring crack-like flaws and (b) on constructing probability distributions of crack depth based on estimated mill rejection rates and inspection quality.
Limit load models and ductile fracture mechanics techniques are employed to identify the governing failure mechanism and a simple unified model is proposed for burst capacity. This model is supported by finite element (F.E.) J-integral analyses and burst tests on P110 and Q125 tubes with artificial axial part-through cracks.
