PhD Oral Exam - Warda Ahmed, Civil Engineering

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

Water reservoirs serve such essential purposes as water supply, irrigation, and hydroelectric power developments. Their effective and safe operations need reliable hydraulic structures for flow control. Siphon spillways are a flow control structure, with two main advantages: 1) no moving parts required and thus less susceptible to breakdown; 2) the ability to pass full discharge with a minimal increase of reservoir water level. The second advantage makes it easier to maintain target water levels as water resources while to ensure safe operations. The design of a siphon spillway uses one discharge head only, termed the design head, which is derived from statistical analysis of hydrological data for the reservoir region. However, under climate change, most existing siphons have been or are expected to be subject to actual discharge heads larger than their design heads. In other words, climate change results in submerged exit condition for siphon flow.

Previous studies of siphons focus on their hydraulic performance under free discharge condition. How submergence affects the flow is a question, not addressed previously. The flow over the crest of a siphon is highly curved. It is a challenge to deal with the curved boundary surface of the crest in siphon analysis. Streamline curvatures, three-dimensional vorticities and other subtle characteristics of the velocity and pressure fields remain to be discovered. Existing studies limit to a qualitative description of siphon flow and oversimplify flow mechanisms. The purpose of this research is to improve our understanding of siphon flow characteristics in a range of submerged exit conditions; to reveal any changes in the hydraulic performance of siphons, compared to free discharge conditions; and ultimately to contribute to an improved design of siphons.

This research took the approach of combining laboratory experiments with mathematical modelling. The experiments used a scaled model siphon. It was fabricated and tested in the Hydraulics Laboratory at Concordia University. A series of experiments were conducted to determine the discharge coefficient in a range of flow rates. A three-dimensional computational fluid dynamics (CFD) model was established for predicting the velocity and pressure fields as well as turbulence quantities. The CFD model used mesh refinement for regions close to solid boundaries and computed two-phase flow. The free surface of reservoir was tracked using the volume-of-fluid (VOF) method. Turbulence closure was obtained using the RNG k‐ε model.

The CFD model predicted detailed distributions of the flow field, including curvilinear flow features in the crest region. The flow characteristics corresponded to various conditions of submergence at the downstream exit of the siphon conduit, and included primary flow velocity, secondary flow velocity, turbulence kinetic energy, and pressure. The predicted discharge coefficient compared well with experimental data. The predicted mean-flow velocities were validated using estimates from the potential flow theory.

The experimental and computational results lead to the following findings: 1) The flow in the upstream reservoir is relatively uniform, with smooth movement toward the siphon entrance. At the entrance, the flow contracts, causing the pressure to drop below the hydrostatic pressure level in the upper corner of the entrance. Under the impact of centrifugal forces, the flow at the crest is forced to move in the extrados direction, causing secondary flow and an increased velocity above the crest. 2) The velocity rapidly increases near the crest and reaches the maximum at a very small height from the crest surface. In the zone above the boundary layer, the velocity decreases gradually before it starts to rapidly decrease near the crown surface. 3) Flow separation occurs just a short distance downstream of the crest. Eddies start to form on the crest and develop along the lower leg of the siphon. Further downstream, the flow contracts by a deflector, creating flow separation on the downstream side. 4) The pressures differ along the siphon conduit, with negative values in the crest region. The pressure decrease is proportional to the increase of velocity. Negative pressures at the crest surface may drop to water vapour pressure at the prototype scale. 5) The discharge coefficient is a parameter of practical importance. It allows one-to-one determination of siphon discharge from the head which is simple to measure. Values of the discharge coefficient range from 0.62 to 0.67 under submerged exit conditions. These values are larger than those under free discharge conditions.

This study has contributed to a better understanding of flow behaviours in a region bounded by curved boundaries. The CFD modelling strategies discussed in this research can be applied to analyse complex flows in similar hydraulic structures. The pressure data reported can possibly be used to access risks of cavitation in siphon spillways. In conclusion, the siphon flow is a complex flow, with curvatures, three dimensionality and turbulence. Through laboratory experiments and mathematical modelling, this research sheds lights on the flow characteristics such as discharge performance, velocity distribution, and pressure distribution.