University of Twente Student Theses


Measuring and modeling the effects of hydraulic and geometrical features of pumping stations on the downtream design of bed proctection

Visser, Thijs (2013) Measuring and modeling the effects of hydraulic and geometrical features of pumping stations on the downtream design of bed proctection.

[img] PDF
Abstract:Bed protection is an important aspect for the stability of hydraulic structures. It protects the bed against large currents inducing bed erosion. The discharge originating from pumping stations can be approximated as a circular turbulent jet. To design a proper bed protection downstream of pumping stations, knowledge is needed regarding the behavior of these jets. Research in the 20th century mapped the behavior of jets in a free environment thoroughly. However, no literature is available on the validity and applicability of free jet behavior in a practical context. Therefore the bed protection is designed with rules that were formulated using scale models in a laboratory setup. As the rules are used to design bed protection behind all hydraulic structures in waterways and hydraulic structures differ largely in details, the design rules are likely to be conservative. Interviews are carried out amongst hydraulic engineers, aiming to determine the practical features and aspects which effects on the velocity field downstream of pumping stations is not clear. The results show that the current design rules provide engineers with too little information on how to take into account the effects of the angle of a valve preventing backflow and the interaction of multiple parallel discharges on the required bed protection dimensions. Also the effect of the outlet orientation with respect to the channel needs further investigation. More insight in these processes will give engineers guidance in including the effects of these features in their design. This will eventually increase an engineer's confidence in his bed protection design. To investigate the effect of these features, first field measurements are carried out near 3 pumping stations in the Netherlands using Acoustic Doppler Current Profiler (ADCP) equipment. The measurements result in a data set containing hydraulic conditions (velocity magnitudes, current directions, discharges) and geometrical conditions (geometry of the channels, the valve angles, the bed profiles). The results show that discharges in length of a small channel tend to attach to one of the banks. When discharging in an angle with the channel, the jet crosses the channel deflecting slightly into the channel. There does not seems to be a collision with the opposing bank. At all structures circulation flow occurs along the edges of the jet, ensuring the conservation of mass. As the valve of the angle cannot be adjusted during the field measurements, the results do not show the effects of the valve angle on the downstream near-bed velocity field. Therefore the structures are modeled using Computational Fluid Dynamics (CFD). The models are validated with the measurement results. The overall performances of the models to represent the velocity magnitudes, expressed in relative deviation, vary largely from 58% to 119%. The relative deviation of velocity magnitudes near the bed, which are more important during this study, vary from 53% to 98%. The large deviations are mainly caused by smaller velocity magnitudes. The model results shows small velocity magnitudes and circulation flow along the banks are both structurally overestimated by the models. The latter causes an underestimation of the dispersion of energy, causing velocity magnitudes further downstream of the structure to be overestimated. Also the jet attachment observed during the measurements is not simulated. During a sensitivity analysis, the number of outlets and the valve angle of the best performing model are varied to investigate their effects on the velocity field near the bed. The results show that the maximum near bed velocity magnitude behind the pumping station is 2.6 times larger at a valve angle of 30° with respect to a valve angle of 90°. The maximum velocity near the bed at a valve angle of 30° is even larger than the depth and width averaged velocity magnitude in the pressure line near the outlet (a factor 1.1). At lower valve angles the jet seems to decelerate faster, and reach lower velocities (<0.5 m/s) sooner than at larger valve angles. The presence of an additional outlet decreases the maximum near-bed velocity magnitude slightly at larger valve angles. For smaller valve angles, the near-bed velocities downstream of the structure are slightly larger (maximal 7%). The results of the sensitivity analysis prove hard to translate into rules for the required bed protection dimensions. The combination of larger maximum near-bed velocity magnitudes and faster deceleration of the near-bed velocity magnitude over distance, makes the formulation of a straightforward relation between the valve angle and the required bed protection length impossible. It is however proved that the improvised rule that takes the valve angle into account, oversizes the bed protection dimensions significantly. The presence of an additional outlet does affect the required bed protection length according to the results. It is found that the improvised design rule, which designs the bed protection length twice as large, is very conservative. Based on the results and their relevance in practical applications, the effect of an extra outlet can be anticipated by lengthening the bed protection with 11% with respect to the single outlet situation.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Subject:56 civil engineering
Programme:Civil Engineering and Management MSc (60026)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page