Long term prediction of pipeline corrosion under tropical seabed sediment

Abstract

The corrosion of pipeline steels buried under seabed sediment is not fully predictable, since there are many parameters affecting the pipeline at different degrees. In relation to that, corrosion growth predictive model based on long term experimental study is of great demand to assist in making suitable pipeline integrity decisions. Therefore, research has been conducted to investigate the metal loss behaviour and corrosion mechanism of pipeline steel under seabed sediment conditions. Then, this study developed a predictive model for corrosion under seabed sediment and highlighted corrosion parameters which presented in tropical climate. Two corrosion models were proposed; one based on the results of a long term exposure of steel coupons to the real field condition. Furthermore, another descriptive model was developed using response surface methodology. The most common applied model used to predict corrosion loss is the power law model (P = ktn), where t is exposure time, and k and n are constant regression of the sediment parameters. Carbon steel coupons were buried in seabed sediment up to two years period. The sediment samples were analysed for its contents and properties. The descriptive model was constructed with aid of Statistica 6.0 software for the data obtained from the laboratory experiment. The results were analysed using statistical methods such as correlation test analysis (CTA), principal component analysis (PCA), multiple linear regression (MLR) and ANOVA (analysis of variance). From the analysis, the extraction of sediment variables related to k and n were successfully obtained. In order to get the best fit of predictive model, the extracted variables are modelled using MLR and embedded in the power law equation. Good curve fitting results are obtained between the actual test data and the proposed models. With consideration of pipelines integrity, the prediction of metal loss due to corrosion in SBS environment using the developed power-law model is considered satisfactory with R2 score of 0.76. The corrosion model based on data from the laboratory has yielded reasonable prediction of metal mass loss with R2 score of 0.83. Noticeably, several sediment factors play an important role in corrosion process and thus determine the corrosion severity. Corrosion growth models have been developed and proposed to predict corrosion progress for steel pipelines buried under seabed sediment. This research has introduced innovative ways to model the corrosion growth for seabed sediment environment. Moreover, intensive statistical analysis has been utilised to determine the level of influence of sediment parameters towards corrosivity. The models enable the prediction of metal mass loss, thus assessing the corrosivity of seabed sediment condition for Malaysian tropical climate.