Mechanical damage is generally considered to be damage that occurs to a pipeline when mechanical excavation, drilling, or boring equipment impinges on a buried pipeline creating scrapes, abrasions, gouges, punctures, and/or dents in the pipeline. Above ground pipelines may also be damaged in a similar manner from impacts by vehicles or projectiles or by wilful acts of vandalism.
In some cases, immediate failure will occur resulting in potentially catastrophic consequences. It is thus important to understand the conditions that would lead to such a failure in order to ensure that design parameters are selected such that immediate failures occur very rarely.
In cases where the damage does not create an immediate failure or the release of gas, the concern generally is that a delayed failure will occur because the strength of the pipeline has been significantly compromised.
In such cases, the possibility is that repeated pressure fluctuations, small increases in pressure, or time-dependent creep will erode whatever margin of safety remains and a failure will ensue. Particularly unsettling are the cases in which damage of this nature is encountered through some form of inspection, where the source of the damage and its time of creation are unknown. In such cases, the operator of the pipeline will generally not know what margin of safety remains.
There are a number of models in existence that may be used to predict both instantaneous and delayed failures due to mechanical damage, and indeed these have been used quite extensively as the basis of repair criteria and for determining safe pipeline operating conditions. Nonetheless, there are significant elements of uncertainty associated with these models and for this reason adequate reserve factors need to be incorporated or recourse must be made to probabilistic approaches that address such uncertainty.
However, since pipelines are getting older and in some cases are being operated at higher pressures than they were previously, there is a requirement to obtain a better understanding of the significance of mechanical damage.In view of this, several research bodies worldwide are taking a keen interest in this topic. To this end, extensive research programs to investigate all key aspects of both delayed and instantaneous failures have been commissioned.
AFAA were commissioned to investigate the conditions that cause instantaneous failures.
The study began by a review of the existing models. It was noted that fit parameters associated with each of the existing models were subject to elements of conjecture, and indeed that various aspects of the models could not be reconciled in the light of modern elastic plastic fracture mechanics theory. In view of this, it was decided to take a step back and to seek a model that could be robustly justified by both theory and available test data.
It was decided to pursue a model based on the failure assessment diagram, which is an alternative means of expressing the ratio of the elastic J-integral to plastic J-integral.
In order to compute the relevant quantities, it was postulated that the gouging and denting processes were likely to result in micro-cracking a sub layer of material adjacent to the surface. Moreover, it was postulated that in addition to level of denting, the extent of the micro cracking was likely to depend on the slip band of the material.
Based on the above postulations a model was derived that incorporated the effects of pressure induced stresses, residual stresses due to denting, a stress concentration factor due to the presence of the gouge, and a stress intensity factor due to the presence of the crack. The model also incorporated a term to determine the depth of the cracking based on the dent depth, the slip-band width, and the grain size.
A high level of agreement between an extensive range of test results was obtained. The model is considered to represent a significant step forward towards the understanding of the behaviour of mechanical damage in pipelines, and provides a sound basis for integrity management strategies.
Development of the model is on-going through additional consideration of the effects of gouging on the stress fields and crack growth. To this end, finite element models using dynamic extremely non-linear analysis techniques have been developed which show good agreement with experimental gouging data.