A Review on Analytic Anomaly Detection of Fluid in Pipeline Using Machine Learning

Fluids such as water, oil, and gas are transported using pipelines in industries. Because of society's social, health, and economic importance, pipelines transporting these fluids should be properly managed. This paper presents a review of the anomaly detection of fluid transported via pipelines using Machine Learning (ML) approaches. We used the conventional method to search, filter, and include relevant papers available in the literature. We classified the included papers based on the fluids category, which includes water, oil, and gas. Numerous researchers propose solutions for detecting anomalies of fluid in pipelines in a generic way. Despite significant contributions from the available works in the literature, we identified the following gaps: lack of research available in the context of water abnormality detection; human and environmental factors are not considered in many works during experiments; lack of research on detection of the anomaly of fluid transported beneath soil.