CO2/H2S Corrosion Rate Prediction Modeling Based on Artificial Neural Network

In order to protect oil-gas pipelines exposed to H2S/CO2-containing environments, a corrosion rate prediction model was built based on the existing corrosion data. A virtual instrument program was developed to realize such prediction using LabVIEW together with MATLAB via the MATLAB Script node. Numerical simulation results indicated that the built model with good stability, high precision and fine effect could provide certain reference for reliability management and predictable maintenance decision in running oil-gas pipelines.