Developing Strategies for the Reduction of Greenhouse Gas Emissions from Wastewater Treatment
Sweetapple, Christine Gillian
Thesis or dissertation
University of Exeter
Reason for embargo
I intend to publish further papers based on the work presented.
This thesis investigates the potential of improved control to reduce greenhouse gas (GHG) emissions resulting from existing wastewater treatment plants (WWTPs), and demonstrates that significant reductions can be achieved without the need for extensive redesign of treatment processes and without increasing operational costs. An emissions model is developed for use in this study, informed by an in-depth analysis of existing state-of-the-art methods and models for estimating GHG emissions, taking into account their suitability for dynamic modelling and WWTP control strategy optimisation. Through the use of local and global sensitivity analysis tools, sources of uncertainty in the modelling of GHG emissions from wastewater treatment are investigated, revealing critical parameters and parameter interactions; these interaction effects have not been considered in previous studies and thus provide a better understanding of WWTP model characterisation. A key finding is that uncertainty in modelled nitrous oxide (N2O) emissions is the primary contributor to uncertainty in total GHG emissions, due largely to the interaction effects of nitrogen conversion modelling parameters. Further local and global sensitivity analysis is used to investigate the effects of adjusting control handle values on GHG emissions, revealing critical control handles and sensitive emission sources for control. This knowledge assists with the following control strategy development and aids an efficient design and optimisation process. Sources with the greatest variance in emissions, and therefore the greatest need to monitor, are also identified. It is found that variance in total emissions is predominantly due to changes in direct N2O emissions and selection of suitable values for wastage flow rate and aeration intensity in the final activated sludge reactor is of key importance. Sets of Pareto optimal operational and control parameter values are derived using a multi-objective genetic algorithm, NSGA-II, with objectives including minimisation of GHG emissions, operational costs and effluent pollutant concentrations, subject to legislative compliance. It is found that multi-objective optimisation can facilitate a significant reduction in GHG emissions without the need for plant redesign or modification of the control strategy layout, but there are trade-offs to consider: most importantly, if operational costs are not to be increased, reduction of GHG emissions is likely to incur an increase in effluent ammonia and total nitrogen concentrations. Alternative control strategies are also investigated and it is concluded that independent control of dissolved oxygen in each aerated activated sludge reactor is beneficial. Optimised solutions are also assessed with respect to their reliability, robustness and resilience, taking into account the effects of influent perturbations and sensor failures on effluent quality and GHG emissions. This reveals that solutions predicted to achieve the most significant reductions in GHG emissions and operational costs under existing design conditions may perform poorly in reality when subject to threats. Dissolved oxygen setpoints which correspond with unacceptable effluent quality reliability and decision variables which should not be considered in future optimisation due to their negative impacts on reliability, robustness and resilience are also identified. Lastly, guidelines for the development of control strategies to reduce GHG emissions are presented. These address GHG emission sources, key control handles and decision variables, choice of control strategy, optimisation and detailed design, and model limitations and uncertainties.
University of Exeter (studentship)
PhD in Engineering