The mad rush to
replace aged, aging and defective pipelines.
Too little can be done for densely populated areas of New Jersey, New
York and other states?
The appraisal of the structural capacity and serviceability
of a pipeline past its design life presents a challenge to many pipeline owners
and operators. While there are many issues that can affect these pipelines,
there are also many techniques that can be used to overcome these challenges.
It is important that those within the pipeline industry share their knowledge
and experiences with aged, ageing or vintage pipelines.
What are the issues?
In particular, pre-1970 liquid and gas pipelines present a
mix of potential problems such as coating deterioration, external/internal
corrosion, stress-corrosion cracking, random mechanical damage (dents and
gouges), fatigue loading (liquid pipelines), pipe weld quality and, although
rare, axial overloading (ground movements). Despite the industry’s good safety
record, it should be emphasised that ‘ageing’ or ‘vintage’ pipelines were
constructed using simple consensus design codes.
The changing legislative requirements, the change in
population density near the pipeline, and the prevention of unstable fracture
propagation (in case crack-arrest capability was not specified at the time of
construction) are other issues to be accounted for. Consequently, the assessment of the condition
of ageing pipelines involves all aspects of temporary pipeline engineering.
What can be done?
Several well-established methods using a deterministic or
probabilistic approach can be applied to assess reported defects/anomalies.
Alternatively, numerical models based on finite-element analyses (FEA) can also
be used. The deterministic or ‘worst-case’ approach can produce overly
conservative predictions, causing undue repairs or requiring operational
conditions lower than those for which the pipeline was designed. With the
probabilistic approach, more-precise predictions are obtained. The predictions
based on numerical models are only as good as the reliability of the input data
and assumptions on which the calculations are based. This issue can be overcome
if the assessment is supplemented with experimental evidence to validate their
use.
However, the uncertainty and the scatter of the input
parameters and the inherent conservatism of the predictions can be difficult to
determine or estimate. For example, the tensile and toughness properties of
many old pipelines are either not available or poorly documented, and if
available, retrieval of relevant data from handwritten material certifications
and construction records might be difficult.
Consequently, to obtain safe predictions, a representative
material property database from which reliable input data can be derived is
needed. Defect-sizing accuracy (ILI data), the defect-interaction criteria
used, and the assessment level or sophistication of the method, have equally a
direct effect on the calculations. Perhaps of greater significance is that
errors in the prediction can be made if unqualified persons conduct the
assessment. Such errors can erode the safety margin. This implies that
technical experience and knowledge of data management are needed to generate
safe solutions.
Depending on the available information, the standard
assessment methods give safe but different answers. For critical applications,
the interaction between the uncertainties can directly be simulated by full-scale
(pressure and or fatigue) testing. Experience shows that this option allows
reduction of the level of conservatism and establishment of a cost-effective
trade-off between the desired operating conditions, inspection capabilities,
and inspection frequency.
The critical issues faced by the
pipeline industry
- ILI and defect-sizing accuracy
- Material properties and test requirements
- Code developments: vintage vs new codes
- Girth welds: inspection and assessment
- Crack-arrest capabilities
- Fatigue testing
- Routing issues, third-party damage
- Coating properties, cathodic protection, corrosion
- Data management
- Risk management.
