| dc.description.abstract | The wastewater sector in the UK and other EU member states are facing stringent regulatory standards. The environmental water quality standards such as the EU-WFD, on the one hand, require a higher level of wastewater treatment which can result in increased GHG emissions and operational cost through higher energy use, chemical consumption, and capital investment. On the other hand, the Carbon Reduction Commitment Energy Efficiency scheme requires the water industries to reduce their GHG emission significantly. The research assesses the advantage of integrated active control of existing WWTPs, their optimisation and dynamic licensing approach to tackle this challenge while maintaining the quality of the receiving river. The dynamic licensing approach focuses on the design of control strategies based on the receiving river’s assimilative capacity. A simulation approach is used to test control strategies and their optimisation, interventions, and dynamic licensing approaches. The study developed an integrated UWWS model that fully integrate WWTP, sewer network, and receiving river, which enables the assessment of the advantage of integrated control strategies and dynamic licensing approach. The hybrid modelling approach uses mechanistic, conceptual and data-driven models in order to reduce computational cost while maintaining the model accuracy. Initially, the WWTP model was set up using average values of model parameters from the literature. However, this did not give a model with good accuracy. Hence, through, a careful design and identification of key parameters, a data campaign was designed to characterise influent wastewater, flow pattern, and biological processes of a real-world case study. The model accuracy was further improved using auto-calibration processes using a sensitivity analysis, identifying influential parameters to which the final effluent and oxidation ditch quality indicators are sensitive to. The sensitivity and auto-calibration were done using statistical measures that compare simulated and measured data points. Nash-Sutcliff coefficient (NSE) and root-mean-square-error (RMSE) measures show consistency in the sensitivity analysis, but correlation coefficient R2 showed a slight difference as it focusses on pattern similarity than values closeness. The combined use of NSE and RMSE gave the best result in model accuracy using fewer generation in the multi-objective optimisation using NSGA-II. Further local sensitivity analysis is used to identify the effect of varying control handles on GHG emissions (as equivalent CO2 emission), operational cost and effluent quality. The GHG emissions both from direct and indirect sources are considered in this study. The indirect GHG emissions consider the major GHG emissions (CO2, N2O, and CH4) associated with the use of electricity, sludge transport, and offsite degradation of sludge and final effluent. Similarly, the direct GHG emissions consider the emission of these major gases from different biological processes within the WWTP such as substrate utilisation, denitrification and biomass decay. This knowledge helps in the development of control strategies by indicating influential control handles and aids the selection of control strategies for optimisation purposes. It is found that multi-objective optimisation can reduce GHG emissions, operational cost while operating under the effluent quality standards. Multi-objective optimisation of control loops coupled with integrated active control of oxygen using final effluent ammonia concentration showed the highest reduction in GHG emissions and reduction in operational cost without violating the current effluent quality standard. Through dynamic licensing approach, the oxygen level in the oxidation ditch is controlled based on the assimilative capacity of the receiving river, which reduces the operational cost and effluent quality index without increased GHG emissions. However, to benefit from the dynamic licensing approach, a trade-off needs to be considered further between final effluent NO3 concentration and reduction in oxygen level in the oxidation ditch to reduce biomass decay which is responsible for higher GHG emission in this scenario. | en_GB |
