Anaerobic Digestion Steady State model parameter estimation for determination of waste activated sludge characteristics

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Phosphorus (P) is an essential nutrient which supports growth and life. A need has developed to recycle P due it being a finite mined resource. At present, most P is lost due to runoff or wastewater (WW) effluent and ending up in rivers and oceans. In order to recycle P and other nutrients, Wastewater Treatment Plants (WWTPs) will need to be converted to Water and Resource Recovery Facilities (WRRFs). However, for WRRFs to be feasible, a better understanding of the current models predicting the fate of P and other material components in place are required. The objective of this study is to utilise augmented batch tests to determine the characteristics of the Waste Activated Sludge (WAS) containing Polyphosphate Accumulating Organisms (PAOs) from a full-scale WWTP as input variables in the Steady State (SS) Anaerobic Digestion (AD) model to ensure accurate prediction of AD performance. The experimental setup used in this research consisted of two completely mixed laboratory scale (20 litres volume, mesophilic 36 ̊C) Anaerobic Digesters (ADrs). The ADrs were operated at SS for 15 day and 32-day Solid Retention Times (SRT) and were fed WAS from a full-scale treatment plant which consisted of a Membrane Bioreactor (MBR) Nitrification-Denitrification (ND) Biological Excess Phosphorus Removal (BEPR) Activated Sludge (AS) system. Two different methods ((i) a novel approach by Maake & Ikumi, (2021) and (ii) the method used by Harding, (2009)) were compared in determining the saturation kinetic rates and the WAS characteristics as input variables for the SS AD models. It was determined that the novel approach by Maake & Ikumi, (2021) was very reliable in determining Chemical Oxygen Demand (COD) removal, Free and Saline Ammonia (FSA), system effluent pH and total alkalinity. With respect to the Ortho-Phosphates (OP), the parent system from where the WAS is sourced, had a long SRT. This resulted in a low predicted PAO count and Poly-Phosphate (PP) content, which resulted in low amounts of Organically bound Phosphate (OrgP) being released into the Anaerobic Digester Liquor (ADL), which equated to an underprediction of OP. Due to the low PP content found in the WAS fed to the ADrs, the ADL was not fully saturated, resulting in no struvite precipitation taking place. This was accurately modelled for both Maake & Ikumi, (2021) and Ikumi, Harding & Ekama, (2013) modelling scenarios.