Detection of Leak Phenomena in Model Pipeline by Characteristics Equation with Viscoelasticity

Water hammer pressure has been widely used for leakage detection in pipeline. This study aims to improve the accuracy of leakage detection by improving numerical simulation of the water hammer pressure. A one-dimensional transient flow model for viscoelastic pipelines is developed by representing the viscoelastic constitutive relationship using the Kelvin–Voigt model and deriving the strain equation based on the transient flow equations, derived from the continuity and momentum equations. A model pipeline is constructed to validate the proposed model. When water hammer pressure is generated in the experimental pipeline, structural vibrations are observed in addition to the water hammer pressure. The signal processing is applied to the measured data to extract the structural vibration, which is superimposed on the water hammer pressure calculated using the method of characteristics to obtain the simulated pressure response. The validity of the proposed simulation is evaluated by comparing the piezometric head data under non-leakage and experimental leakage conditions. Under leakage conditions, the experimental data agrees well with the simulated results, demonstrating that the proposed simulation, which combines the viscoelastic transient flow model with the extracted structural vibration component, is accurate and reliable. In conclusion, this study proposes a simulation method that considers not only pipe-wall viscoelasticity but also the experimentally observed structural vibration. The proposed method successfully reproduces the experimental results under leakage conditions and demonstrates the potential to reproduce responses that are difficult to represent using the conventional viscoelastic model alone. These findings indicate that the proposed method has the potential to improve the accuracy of water hammer simulations and can contribute to the development of more accurate leakage detection techniques for pipelines.