Prediction of wastewater treatment greenhouse gas using a real-time model: PAPER & PRESENTATION
Steve Bungay, Technical Director & Anna Whitmore, Senior Digital Consultant, Mott MacDonald, UK
(free)
Nitrous oxide (N2O) is the major contributor to greenhouse gas (GHG) emissions from wastewater treatment plants. However, the underlying mechanisms leading to its production are complex and not fully understood.
At present, fixed emission factors (EF) are typically used to estimate N2O emissions. However, these factors do not reflect the level of uncertainty around how emissions are affected by several influences, including wastewater characteristics, the type of treatment process, and how the process is operated and controlled. Whilst mechanistic models have been developed for predicting GHG emissions, these are not yet well proven for predicting and reporting emissions. Additionally, these models typically represent a fixed point in time, and don’t benefit from real-time data inputs and control opportunities.
The Rosedale Wastewater Treatment Plant (WWTP), Auckland, achieves biological nitrogen removal through nitrification and denitrification. As with all biological treatment plants, the operation results in direct emissions of greenhouse gases such as CO2, CH4, and N2O. In 2020, a digital twin was developed to provide real-time insights into plant performance, and live reporting of GHG emissions. The objective was to assist with planning and benchmarking and to help achieve the aim of reducing operational process emissions by 50% by 2030. A combination of off-gas and dissolved N2O was also analysed to provide information on where in the process N2O was generated. Real-time data was captured in SCADA and surfaced to a dashboard in the digital twin. The digital twin will provide ongoing real-time predictions of GHG emissions in the absence of permanent off-gas analysers, allowing these to be redeployed to other sites, where similar models can be established.
This paper documents the development of the model, the practical challenges of collecting and interpreting measurement data, discuss the key factors influencing the production of GHG emissions in the Rosedale plant, and the reliability of the real-time model for predicting emissions.
