A Two-Stage Deep-Learning Based Detection Method for Pipeline Leakage and Transient Conditions

Nowadays, Leakage detection is of great importance as pipelines are the major means of transporting hydrocarbon fluids and gases. In this paper, a novel two-stage detection method is introduced to differentiate normal, leakage and transient conditions of pipelines. In this method, feature vectors are constructed from the flow difference and pressure using leakage characteristics, and are normalized with the modified hyperbolic-tangent estimator. An artificial neural network is used in the first stage of detection to differentiate normal and abnormal conditions with the feature vectors as the inputs. In the second detection stage, a simple logic is used to distinguish leakage and transient from abnormal time-windows. In addition, a pre-set leak-size tolerance is used to trigger alarms for detected leakage time-windows. The results for the cases of using different machine learning methods and varying leak-size tolerances are given. The method has been shown to have higher detection performance and less false alarms in comparison with the line balance and Kantorovich distance methods.