Predicting Dissolved Oxygen and Nitrogen Uptake During Turbine Aeration

摘要

During hydropower production, stratification of the upper reservoir can result inrnexceedingly low levels of dissolved oxygen (DO) a distance below the water surface. Inrnmany sites, water from these depths enters the turbine intakes where it is transported tornthe tailwaters downstream of the plant. There, low levels of dissolved oxygen can havernan adverse effect on aquatic life and water quality. In an effort to reduce thernenvironmental impact of hydropower generation, governmental regulations are beingrnestablished which require minimum tailwater DO concentrations.rnA variety of techniques have been employed to establish the air-water mixturernnecessary for dissolved oxygen enhancement of the water exiting the turbine, includingrna) line diffusers bubbling pure oxygen into the upper reservoir; b) venting air into thernturbine; and c) spilling or utilization of tailrace weirs. In many plants, turbine venting is arnparticularly cost effective method for increasing tailrace DO levels by introducing air intornthe water passing through the turbine. There, the concentration gradient between therntwo phases causes oxygen and nitrogen in the gaseous state to dissolve into thernsurrounding water. Despite the benefits of DO enhancement, the accompanyingrnincrease in total dissolved gas (TDG) can be harmful to water quality and must bernevaluated in conjunction with the DO increase.rnRecently, the discrete bubble model (DBM) methodology was developed tornpredict mass transfer in the air-water mixture occurring within airlift aerators, the SpeecernCone, bubble-plume diffusers and turbines. While the DBM is sensitive to turbine andrnplant geometry, as well as discharge rate, the predictions are dependant on initial bubblernsize at the injection location and bubble dynamics during transit. At these early stages ofrndevelopment, correlation with test data has been solely based on average dischargernvelocity. However, complex flow patterns within the turbine, such as the part load ropernvortex, cause local velocity gradients that will influence bubble size and mass transfer.rnThese flow patterns, and the presence of admitted air, will also dictate the amount of airrnthat can be drawn into the turbine by influencing the air inlet pressure. Insight into howrnthese flow characteristics affect bubble size and air flow can be obtained throughrncorrelations based on the ratio of discharge rate Q to the rate at which peak efficiencyrnoccurs, Q_(opt), i.e., Q/Q_(opt). After establishing these relationships for distributed, central andrnperipheral aeration, the current paper presents a comprehensive calculationrnmethodology that incorporates individual turbine design and operation into air flowrnpredictions and the corresponding dissolved oxygen and nitrogen uptake within therndischarge.