Leak diagnosis in pipelines using a combined artificial neural network approach

Leakages in pipelines affect the reliability of fluid transport systems causing environmental damages, economic losses, and pressure reduction at delivery points. Therefore, this paper presents a methodology to detect locate water leaks by using artificial neural networks (ANN) techniques online measurements flow rate. Contrary reported works literature, proposed method estimates friction factor pipe uses information as an input compute leak position. Data generated from validated numerical simulator was used enrich data-training set for ANN. Various scenarios were considered characterize losses their differentials different sections pipeline. Finally, algorithm tested experimentally pilot plant. The results demonstrate good performance applicability method.