The key point of pipeline integrity management with regard to external corrosion is the control of the corrosion activity at coating defects. In practice pipeline integrity management often results in the execution of intensive field surveys and monitoring programs leading to high operational cost.
The general approach is to apply external corrosion direct assessment techniques and to use in-line inspection tools to gather as much information as possible on the corrosion status. Correlating different field data like wall thickness measurements, pipe-to-soil potentials, soil characteristics and coating conditions for predicting the corrosion risk is a challenging task. Moreover corrosion assessment is very complex in congested areas or where AC/DC interference occurs. Understanding and simulating the corrosion mechanisms allow a correct and timely analysis which is more difficult to achieve with statistical tools.
This article presents a new approach in the pipeline integrity management which is based on mechanistic modelling. The electrochemical reactions taking place at coating defects are simulated for the entire pipeline network, even in the presence of AC and DC interference resulting in the visualization of the IR-free potentials and corrosion rates.
Pipeline corrosion is perfectly preventable using techniques such as protective coatings, cathodic protection (CP) and drainage systems against AC/DC interference. Coating degradation and anode consumption decrease corrosion protection efficiency over time. It is therefore crucial that any preventive measures be monitored to guarantee their unremitting effectiveness. But to safeguard the integrity of pipeline networks, proactive management and timely identification of corrosion threats count most - especially in the case of interference from third-party systems.
The conventional approach for managing external corrosion threats is to apply external corrosion direct assessment (ECDA) techniques such as direct current voltage gradient (DCVG), close interval potential surveys (CIPS), and alternating current voltage gradient (ACVG), etc as described in the ASME B31.8. 1] The amount of data is very large and difficult to interpret especially when the pipeline is close to other buried infrastructure. Statistical models such as bow-tie and Baysian models help finding correlations in the field data set but provide rather qualitative than quantitative results. The statistical approach can predict the root cause of some failure modes if the size of data related to it is sufficiently large for obtaining a high probability of the predicted outcome. In other words, the variance in data should be sufficient small to draw proper conclusions, which is relatively difficult to obtain in complex pipeline systems such as multiple pipeline corridors in proximity of high-voltage power lines. Besides general trending, concerns of higher importance involve sudden changes, or atypical corrosion/cathodic protection (CP) behavior that is observed during the daily operations or immediately after a survey campaign. In some cases, prioritization needs to be elevated and immediate action is required, particularly when interference is the origin of this behavior.