Date of Award

Summer 2019

Document Type

Open Access Dissertation


Civil and Environmental Engineering

First Advisor

M. Hanif Chaudhry


Piping systems are commonly designed to withstand the first transient pressure peak, which is unaffected by dissipation. However, for multiple operations of control equipment, e.g., pump start-up following pump shutdown, load acceptance following load rejection on hydraulic turbines, etc., an accurate prediction of the dissipation of pressure oscillations is needed to determine a suitable time for the second operation. For this purpose, a method is presented to compute the dissipation of pressure oscillations in piping system following a simple procedure used for computing the dissipation of vibrations of bridges and other structures. Similar to structural engineering, this method is simple to apply, does not require simulation of the entire system, is not computationally intensive, and gives reasonable results for practical applications for a complex phenomenon mechanics of which is not well understood. An empirical equation for the damping ratio is developed by using the dimensional analysis and by nonlinear regression. Comparisons of the computed and experimental results for 17 tests conducted in laboratories all over the globe show good agreement. It is found that the damping ratio increases with increase in Reynolds number or Mach number but decreases with diameter to length ratio of the pipeline. The uncertainty of the model to determine the damping of pressure head oscillations following a sudden valve closure in a simple piping system in pressurized closed conduit using Bayesian inference is quantified. The joint probability density of the model parameter is estimated based on experimental results published in the literature as well as from experiments performed at the University of South Carolina. A Markov Chain of the posterior joint distribution of the model parameters is calculated and used to predict the pressure head oscillations. The prediction is based on a probabilistic analysis to estimate an interval of pressure as a function of time, rather than estimating a single point if simple regression analysis is done. The 95% high posterior density of the damping ratio is found to range between 1% to 6%. Uncertainty analysis shows that the model predicts the value of the damping ratio successfully.