Development of Mechanistic Models and Numerical Methods for Corrosion Prediction in Oil Industry
Abstract
Corrosion is the primary mechanism resulting in facility failures in oil industry, where downhole tubulars, above-ground gathering pipelines and buried transmission pipelines constitute the important facilities. Corrosion modelling is advantageous over experimental studies, and has been essential for industrial corrosion management.
Internal corrosion of pipelines occurs in CO2-containing single-phase water or oil-water emulsion flow due to dissolved CO2 in water. Corrosion of X65 pipeline steel under various flow conditions is studied with a home-made flow loop. A semi-empirical model is developed based on computational fluid dynamic (CFD) simulations to predict the corrosion rate of the pipe steel in CO2-saturated oil-water emulsion flow.
The high-temperature high-pressure conditions encountered in steam-assisted gravity drainage (SAGD)/CO2 co-injection systems introduce major corrosion concerns to downhole tubulars. A semi-empirical mode is developed to predict the tubular corrosion rate under the SAGD/CO2 co-injection conditions. The corrosion rate is very small when a compact scale is formed.
The CO2 storage is used worldwide to help reduce CO2 emission. However, supercritical CO2 conditions in sites can lead to severe corrosion to steel tubing. To predict corrosion rate of the steel tubing under CO2 storage conditions, a mechanistic model, which includes a water chemistry sub-model and an electrochemical corrosion sub-model, is developed. The predicted solution pH and corrosion rates are well consistent with the experimental results.
External corrosion is the dominant threat to structural integrity of abandoned pipelines in soils. Studies of corrosion of X52 pipeline steel in a simulated Regina soil solution show that the iron oxidation and oxygen or water reduction are electrochemical anodic and cathodic reactions, respectively, depending on the dissolved oxygen level. Porous corrosion products reduce somewhat the corrosion rate. A mechanistic model enabling prediction of the long-term corrosion rate is developed. Corrosion kinetic parameters used in the model are obtained by fitting the experimentally measured polarization curves with a computer program. The model is validated by comparing the calculated results with experimental data.
Description
Keywords
Engineering--Chemical, Materials Science
Citation
Li, Q. J. (2017). Development of Mechanistic Models and Numerical Methods for Corrosion Prediction in Oil Industry (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/25681