Prediction of submarine pipeline corrosion based on parameter optimized GM-Markov model

To avoid accidents such as perforation leakage caused by submarine pipeline corrosion, and to timely maintain and protect submarine pipelines, the prediction method for submarine pipeline residual life based on parameter-optimized GM-Markov model was proposed in combination with the characteristics of traditional grey system to process little data, poor data and Markov theory to predict the future state. The feasibility of building the grey GM (1, 1) model was firstly analyzed, then the GM (1, 1) model with optimized parameters was built, and the initial conditions of the model were changed to predict the corrosion depth of submarine pipelines. According to the predicted corrosion depth, Markov model was used to quantitatively analyze the future corrosion state of submarine pipelines and predict their residual life. Some submarine pipeline test section was selected as an example to predict the residual life of the pipeline. The results show that the model with the initial conditions changed enables higher prediction accuracy with only a small amount of sample data, demonstrating its worth of popularization and application.