CFD Modeling of Air Pocket Transport in Conjunction with Spillway Conduits

This paper focuses on simulations of enclosed air pocket movements in conjunction with bottom outlet operations. The critical velocity of water for air pocket transport in pipe is the minimal flow velocity for the air pocket start to move downstream. A numerical model is developed to simulate the critical velocity of air pocket transport in pipe flow and to discuss the impacts of tunnel slope, size of the air pocket and wall roughness. The computations are performed in FLUENT using Volume of Fraction (VOF) model combined with kepsilon model. Parallel computing is adopted for high computational performance. The modeled critical velocity is compared with experimental results and they increase with increasing slopes. However, as the roughness height defined in the model is not big enough to represent the reality and no wall shear stress is applied in the upper wall where air pocket and wall contact, the modeled critical velocity is smaller than the experimental ones. Therefore, wall roughness contributes to keep the air pocket from moving downstream which is important in modeling critical velocity. However, by assuming a constant wall shear stress for the air phase the same as the water phase will overestimate the shear stress on the air pocket. Two air pocket volumes are simulated at the slope 0.8 degrees which shows the bigger the air pocket is the higher the critical velocity is. Modeling results also show that the critical velocity is non-zero in horizontal pipe and there is a limit for the carrying capacity at all slopes. The simulations of air pockets with different volumes in the bottom tunnel of Letten dam in North of Sweden is shown in this paper as well. INTRODUCTION The bottom outlet of a dam is used for emptying the reservoir, flushing sediment and discharge diversion in addition to the surface spillways [1]. Figure 1 shows a type of bottom outlet in Sweden. It is obligatory to have bottom outlet as part of the design in many countries for safety concerns. Many of the Swedish dams have bottom outlets. However, most of them are not in use after construction, partially because entrained air in pressurized flow in the tunnel could cause problems during operation. The problems can be many: the discharge capacity of bottom outlets at full gate opening can be greater than what was originally designed; vibration of the gates/conduits and pulsations at the discharge to the downstream could happen; cavitation in the conduits is also one of the problems. All of these would harm the safety operation of the bottom outlet and cause serious consequences. Therefore, the operational safety must be guaranteed without unexpected risks and operating conditions requires the reliability of bottom outlets. According to the inventory of bottom outlets in Sweden, bottom outlets of 38 dams were investigated [2]. The discharge capacity of bottom outlets is determined with the help of model tests or computations. Around half of these 38 bottom outlets were model-tested. The problem with computations to find out the discharge capacity is that the calculations are correct at maximal or full flow. Otherwise, problems associated with partially pressurized flow would occur, leading to numerically incorrect results. As shown in Table 1, 8 out of 9 detected problems during the operations of bottom outlets are directly related to air. This makes the study of the air entrainment in the bottom outlets and interesting issue in order to ensure the safety statues of outlet functions. Table 1: Reported problems in bottom outlets [2] Type A B C D E Other Total Reported problem 1 4 1 1 2 0 9 Reported problem directly related to air 1 4 1 0 2 0 8 Type A: rock tunnel bottom outlet, Type B: shaft bottom outlet with partial pressurized water way, Type C: bottom outlet culvert, Type D bottom outlet without or with very short waterway, Type E: combined surface and bottom outlet.