Alternative Methods for Evaluation of Oxygen Transfer Performance in Clean Water

Abstract: Aeration of wastewater is performed in many wastewater treatment plants to supply oxygen to microorganisms. To evaluate the performance of a single aerator or an aeration system, there is a standard method for oxygen transfer measurements in clean water used today. The method includes a model that describes the aeration process and the model parameters could be estimated using nonlinear regression. The model is a simplified description of the oxygen transfer which could possibly result in performance results that are not accurate. That is why many have tried to describe the aeration at other ways and with other parameters. The focus of this Master Thesis has been to develop alternative models which better describe the aeration that could result in more accurate performance results. Data for model evaluations have been measured in two different tanks with various numbers of aerators. Five alternative methods containing new models for oxygen transfer evaluation have been studied in this thesis. The model in method nr 1 assumes that the oxygen transfer is different depending on where in a tank the dissolved oxygen concentration is measured. It is assumed to be faster in a water volume containing air bubbles. The size of the water volumes and the mixing between them can be described as model parameters and also estimated. The model was evaluated with measured data from the two different aeration systems where the water mixing was relatively big which resulted in that the model assumed that the whole water volume contained air bubbles. After evaluating the results, the model was considered to maybe be useful for aeration systems where the mixing of the water volumes was relatively small in comparison to the total water volume. However, the method should be further studied to evaluate its usability. Method nr 2 contained a model with two separate model parameter, one for the oxygen transfer for the air bubbles and one for the oxygen transfer at the water surface. The model appeared to be sensitive for which initial guesses that was used for the estimated parameters and it was assumed to reduce the model’s usability. Model nr 3 considered that the dissolved oxygen equilibrium concentration in water is depth dependent and was assumed to increase with increasing water depth. Also this model assumed that the oxygen was transferred from both the air bubbles and at the water surface. The model was considered to be useful but further investigations about whether the saturation concentrations should be constant or vary with water depth should be performed. The other two methods contained models that were combinations of the previous mentioned model approaches but was considered to not be useful.