Affordable and Clean Energy (SDG 7), Responsible Consumption and Production (SDG 12)
Modelling corrosion behaviour and designing high performance corrosion resistant steel for oil pipelines in South Africa using Artificial, Convolutional, and Recurrent Neural Networks (ANN) Matsolo Seloanyane 1 *, Dikanua Nkazi 2 , Biltayib Misbah Biltayib 3 1 School of Chemical and Metallurgical Engineering, Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, South Africa 2 Assistant Professor, School of Chemical and Metallurgical Engineering, Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, South Africa 3 Assistant Professor, Petroleum Engineering College of Engineering, Australian University, Kuwait Internal corrosion of the steel pipelines is a major challenge within the petroleum industry, resulting in damages estimated at $2.2 trillion and amounting to 3 – 4 % of the world’s GDP 1 . It is the main cause of deterioration and degradation in the pipelines, and it is detrimental to the people and the environment causing leakages, spillages, and explosions 2 . The corrosion behaviour of steel is a complex phenomenon, and traditional corrosion modelling techniques have limitations in accurately predicting corrosion. In this study, artificial, convolutional, and recurrent neural networks, will be used to model the corrosion behaviour of steel, based on various factors such as pH, temperature, and chemical composition. These algorithms will be used to develop a corrosion prediction model that will predict the internal corrosion rate profile within oil and gas pipelines and determine which sections of the pipeline is corrosion most likely to occur. Real field data will be used to train, test, and validate the model. The results of the model will be used to design a high-performance, corrosion-resistant steel for oil and gas pipelines. This research is important because it will provide a better understanding of the corrosion behaviour of steel and help to develop more durable and reliable pipelines. The results will also help in the pipeline manufacturing process where expensive failure repairs could potentially be avoided by correcting these problems early on in production. References 1. Bardal, E., 2007. Corrosion and Protection . Springer Science & Business Media. 2. El Ibrahimi, B., Jmiai, A., Bazzi, L., El Issami, S., 2017. Amino acids and their derivatives as corrosion inhibitors for metals and alloys . Arab J. Chem. 3. Enani, J., 2016. Corrosion control in oil and gas pipelines . Int. J. Sci. Eng. Res. 7, 1161–1164.
P20
© The Author(s), 2023
Made with FlippingBook Learn more on our blog