Optimization methods for sedimentation and denitrification in activated sludge

Abstract: Denitrification on readily biodegradable carbon compounds and separation of biological sludge from effluent water are examples of critical sub-processes for nitrogen removal in activated sludge. The aim of this thesis was to find operational strategies and newmethods to optimize these processes. A simple method to measure the content of readily biodegradable organic matter in wastewater is presented in the thesis. Also a work on intermittent addition of carbon to a pre-denitrifying system is included. In this work it was shown that despite regular 5-day interruptions in ethanol supply, the bacteria maintained a sufficiently high denitrification capacity. The second and major part of the thesis deals with separation of biological sludge from effluent water. It includes two methods to decide and monitor the limiting solids flux; firstly by comparing the curvesof solids flux and sludge blanket heights visually and secondly by comparing real and simulated sludge blanket height curves. The simulated curve is predicted by a recursively estimated first order linear difference equation model. The second part of thethesis also include a calibration method for one-dimensional sludge settling models, to be used for simulations of sedimentation performance in a rectangular clarifier. Results from simulations agreed well with full-scale performance. Finally, a statistical analysisof variations in sludge settling capacity showed that a single floc characteristic parameter value could only explain about 15% of the variation. With the use of multivariate statistics, 45% of the variation could, with very high significance, be explained by the parameters total extended filament length, floc compactness andextracellular polymer substance abundance.