In this study, a three-dimensional two-phase (air and water) numerical solver is applied to investigate free surface flows. The first component aims to improve the overall understanding of the underlying physical mechanisms that occur during the interaction between turbulent hydraulic bores and simple structures. Data collected during large-scale physical experiments based on generating dam-break waves in a horizontal rectangular channel is used for comparing to the numerical results. An extensive sensitivity analysis on numerical parameters including spatial discretization and turbulence models is presented. Quantitative comparisons of numerical and experimental time series of water surface elevations, pressure, and net streamwise force exerted on the structure are used to validate the model. In the in-depth analysis, it is demonstrated that the model is able to simulate the pertinent aspects of the flow behaviour that occur during the interaction with good agreement. The numerical impulsive force generated at initial impact shows excellent agreement with the experimental results, particularly for the larger magnitudes bores considered. Since the numerical model treats the air as an incompressible media, the level of agreement observed between the experimental and numerical results suggests that the compressibility of the air in the leading edge of the bore during the physical testing had no significant effect on the measured impulsive force. The two-phase model was also able to capture the occurrence of a second transient spike in the force exerted on the structure when the initial runup collapsed back onto the incoming flow, trapping a pocket of air in the process. The model was further applied to investigate the effect of an initially quiescent layer of water in the downstream channel section on bore propagation characteristics and the subsequent interaction with the structure. It is demonstrated that for small nonzero values of initial downstream depth a substantial increase in bore depth occurs. However, further increases in the downstream depth did not appear have any significant effects. For the greatest downstream depth simulated, a considerable reduction in the hydrodynamic force is observed as a result of a more rapid closing of the wake that develops on the leeside of the structure.
The second component of the study applies the same numerical solver to investigate a novel long wave generation technique for producing laboratory-scale tsunami waves. The concept is based on removing the air from the inside of a tank with a submerged outlet at the upstream end of the basin and releasing the water in a controlled manner. A similar procedure as described above was used to calibrate the numerical parameters to experimentally-measured wave heights and periods. To model the influence of the pneumatic valves mounted on top of the upstream chamber, time-varying pressure boundary conditions are developed to regulate and control the pressure inside the tank. Quantitative and qualitative comparisons of the numerical and experimental results show good agreement and a high potential for the solver to be used for similar investigations. An analysis is performed to improve the existing understanding of the wave formation process. The model is also applied to modify test configurations that influence the waveform for which the results may be used to aid in making operating decisions for future tests or in the design of similar wave generating devices.
| Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/34076 |
| Date | January 2016 |
| Creators | Douglas, Steven |
| Contributors | Nistor, Ioan |
| Publisher | Université d'Ottawa / University of Ottawa |
| Source Sets | Université d’Ottawa |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0016 seconds