Whether needed for transporting oil, gas, hydrogen, carbon dioxide, or water, pipelines are being designed for and installed in increasingly deeper waters; water depths of more than 2,000 m are no longer an exception. The main driver for the design of such pipelines is their resistance against hydrostatic pressure, especially during installation. Previous studies by the authors culminated in an improved method for predicting the collapse pressure. The authors compared their analytical results with the outcome of finite-element analyses. In this paper, the performance of the improved model and the conventional DNV model, DNV, is compared with the results of full-scale collapse testing. The predictions of the improved model are significantly closer to the actual test results of cold-formed pipe in an as-fabricated condition compared with the conventional model’s predictions. For a pipe subjected to (light) heat treatment, both methods’ estimates are relatively close to the test results. The Bauschinger effect and the associated change of the stress–strain curve’s shape must be reflected in the model to achieve consistent and accurate results.

This content is only available via PDF.
You can access this article if you purchase or spend a download.