MACHINE LEARNING DRIVEN DEVELOPMENT OF CORROSION INHIBITORS IN OIL AND GAS INDUSTRY APPLICATIONS

Abstract

Corrosion inhibitors remain one of the most widely used and effective strategies for mitigating corrosion in the oil and gas industry. This study employed a machine learning (ML)-driven approach to develop green corrosion inhibitors, utilizing a Graph Convolutional Network (GCN) to predict inhibition efficiencies. The model was trained on a dataset of over 100 inhibitors and predicted an inhibition efficiency of 84% for 200 ppm chitosan-grafted polyacrylamide (CsAM). Experimental validation was conducted using electrochemical techniques on CsAM, synthesized with four different polyacrylamide-tochitosan ratios. Characterization of the inhibitors was performed using Scanning Electron Microscopy (SEM), Fourier Transform Infrared Spectroscopy (FTIR), Contact Angle measurements, and Thermogravimetric Analysis (TGA). Electrochemical results showed a maximum inhibition efficiency of 98% for 1:30 CsAM at 200 ppm. Corrosion kinetic analysis revealed that the inhibitor acts as a mixed-type inhibitor, with corrosion prevention primarily governed by physisorption.