Detection of faults in subsea pipelines by flow monitoring with regression supervised machine learning

Abstract

This study investigates the relationship between pressure change, velocity change, and temperature of crude oil through a pipeline and presents a method of using a regression supervised machine learning (ML) algorithm to detect faults. A representative dataset of crude oil flow is generated by computational fluid dynamics (CFD) and used to train the algorithm to develop a model of fluid behavior under normal pipeline operations over a range of typical flow rates and temperatures. CFD data are then collected under several simulated fault conditions: leaks of 10% and 20%, and a 50% restriction to flow, by nominal pipe cross-sectional area. This study demonstrates that the ML algorithm can be trained to model the system under normal conditions, thereby successfully recognizing a fault condition as non-conforming and indicative of a statistically significant change in pipeline operation. It is further able to identify the fault type based on the pattern observed in the new data. It is shown that ML may be a safe, low-cost, and accurate method of monitoring a subsea pipeline for optimal performance and fault detection without the need to introduce special equipment to a subsea pipeline network, providing an avenue for enhanced process safety and protection of ocean environments. This paper demonstrates that the application of ML to the monitoring of pipeline networks could provide valuable contributions to the industry in terms of safety, cost, and environmental protection.