Effect of 3D random pitting defects on the collapse pressure of pipe - Part II: Numerical analysis

In the second part of this investigation, a 3D nonlinear (Finite Element) FE models was established to systematically study the collapse pressure of random corroded pipe. The models are developed within the framework of the nonlinear finite element ABAQUS in combination with a user defined Python program. Two constraints technique including surface-based tie constrain (SBTC) and shell-to-solid coupling (SSC) were implemented to take the complex geometry into account without the sacrifice of model accuracy while minimizing the computational cost. It was found that SBTC is more robust to simulate the locally random corroded pipe (external corrosion) than SSC. After validation models, a parametric investigation is performed to study the influence of wave number and amplitude of out-of-roundness, random pitting factors (pitting shape, pitting density and the length of pitting affect zone (LC)). It was revealed that the amplitude (wmax) and wave number (n) of out-of-roundness, pitting density and mass loss are the main factors that govern the collapse pressure (PCO) of 3D random corroded pipe whereas pitting shape has minor impact on PCO. Finally, a simplified pitting model was proposed, it was found that the circumferential and axial bending rigidity are key parameters that govern the PCO of pipe, which is mainly determined by the minimum remaining area in circumferential and axial direction.