The Prediction of External Corrosion in Buried Pipelines from Survey Data
By C Lee and F Arteche
Industrial Need
All carbon steel underground pipelines are subject to corrosion where the protective coating is damaged and there are inadequate levels of cathodic protection. The issue is particularly significant in ageing pipelines. A commonly used approach for the assessment of external corrosion risk of buried, land pipelines is based on the NACE RP502 standard, often referred to simply as “Direct Assessment”, or External Corrosion Direct Assessment (ECDA).
This report describes work carried out to provide a better understanding of predicting external corrosion on buried pipelines and is based on a case study associated with an integrity assessment carried out by TWI on pipelines using ECDA.
Key findings
- The Direct Current Voltage Gradient (DCVG) %IR value obtained from indirect inspection is found to correlate well with the actual coating defect size, but the relationship is affected by other factors, i.e. linear regression did not show a good correlation but a clearer correlation was shown in multiple linear regression
-
Quantile regression has been found to be a useful tool to understand the influence of %IR for small and large coating defects compared with the simplicity of multiple regression models.
-
The correlation of %IR value to coating defect size is non-linear with sensitivity increasing as the %IR value increases, i.e. larger %IR values gave, proportionally, a larger coating defect area.
-
The DCVG %IR value obtained from indirect inspection does not correlate with corrosion depth.
-
The actual coating defect size does not correlate with the depth of corrosion found; the likelihoods of corrosion occurring at small defects and also at large coating defects are similar.
Excavated coating defect location on a coal-tar coated pipe
Excavation at a coating defect location showing corrosion on the pipeline after cleaning
