Turning Up the Heat on Thermal Modeling: Advancements in Thermal Modeling at Resource Recovery Facilities

Background
The National Pollution Discharge Elimination System requires that discharges into Waters of the United States do not degrade the receiving water body. Much attention has been given over the past half century to biological oxygen demand and nutrient loading; however, over the past two decades, research has shown growing negative impacts of increased water temperatures on native species in several US regions [e.g. Coho Salmon within the Pacific Northwest]. While there are several ways of reducing the excess thermal load of municipal resource recovery facilities on receiving water bodies, the majority of the programs where large scale thermal load reduction has been required and implemented are either reactive (ie implemented through permit renewals for existing facilities via more stringent thermal load limits) or are for new industrial discharges, where the anticipated thermal load is typically driven by known processes making it more predictable. While a reactive approach has worked thus far, with climate change and urban population growth set to make the thermal loads on water bodies all the more impactful, there is a need to better understand and predict the thermal load of municipal resource recovery facility effluent streams. Not only will this allow for a more proactive approach to thermal management, a better understanding of the thermal sinks and sources within a plant provides an opportunity for low effort optimizations to reduce thermal loads and helps create better predictions of biological treatment processes which can suffer from reduced efficacy at lower process temperatures.
Objectives
In support of evaluations for a new municipal treatment plant and discharge in the Pacific Northwest, this team conducted in depth 'whole system' thermal modeling to determine the cooling needed to meet permit requirements for a new MBR plant both at initial commissioning in the mid-2020s and at various stages of buildout, up to and beyond 2040. This predictive temperature model was created using Excel, Biowin, SWMM, and historical environmental data from the recently expanded National Energy Research Laboratory (NREL) GIS climate data repository. This allowed for predictions of the effluent temperature throughout each year of the system buildout. This model accounts for the following. •Heat flux from external sources such as, Solar and Atmospheric Radiation, Aeration, Conduction and Convection, Evaporation, Mechanical equipment, and Biological reactions. •Advective and Diffusive mixing. •The configuration of the process basins (ie FRP or concrete covers; footprints and depths; primary, recycle, and secondary process flows, etc) Furthermore, by using the NREL environmental data, the model is able to calculate the effective cooling of various active and passive cooling methods accounting for many variables that make performance prediction more difficult. These variables include humidity, ambient air temperature, and air speed. For the purpose of this evaluation the anticipated thermal cooling capacities of cooling towers, chillers, surface aerators, blower aftercoolers, and effluent pipeline geothermal cooling were also able to be evaluated.
Methodology
Daily plant influent temperature data from the last two years, the NREL environmental data from the past 5 years and sampled at 30-minute intervals, Biowin process model results at various flows and process temperatures, and SWMM collection system data predicting collection system flows for the next 20 years where combined into a database that was able to be accessed in excel and interpolated using several lookup functions. As such this was able to provide access to trends over the last five years and up to 20 years in the future. Once this data was made available in excel it was pulled into other parts of the model. Using the above environmental and process data, a flow model was built to account for the total process volume in each basin of the treatment plant and the associated flows between all of the basins – this allowed for the heating and cooling effects in each cell of the wastewater treatment facility to be determined as detailed in the above approach section. This portion of the model was built on the strategies evaluated and proven in the below research, with certain improvements made to account for the increased accuracy and frequency of environmental data.
-'Temperature Modeling in Activated Sluge Systems: A Case Study' by Jacek Makinia, Scott A. Wells, and Piotr Zima.
-'Temperature Modelling and Prediction for Activated Sludge Systems' by Silvia Lippa, Roberto Canziani, Diego Rosso, and Michael Knudson Stenstrom
-'Pulp and Paper: Temperature Modeling in Large Basins' by Marotta and Pezeshk.
Using Visual Basic for Applications an iterative calculation engine was created that allowed the thermal model to progress in 1 minute time intervals. The the current thermal fluxes and transfer rates every 2 hours were used for further analysis, trending, and continuation of a simulation as needed.
Calibration
The model has a high level of anticipated accuracy based on past academic research such as Makinia et al, 2004 and the other sources noted above. However, current phases of this evaluation will perform additional validation via high sample rate (1 sampler / 5 minutes), high accuracy (+/- 0.01 degF), long term temperature monitoring performed at two facilities: (1) a nearby treatment facility with a different process scheme and (2) a treatment plant located farther away but with the same process scheme. While validation is occurring one set of temperature
Findings
The thermal modeling determined that the treatment facility will need cooling during certain months of the year to maintain permit conditions; however, it identified that up to a 30% reduction in the thermal load could be achieved through small changes to the plant design and optimization of the plant discharge pipeline. Additionally, past thermal models preformed on conventional activated sludge treatment systems showed that the thermal load from biological activity represented 25-30% of total heat gain; however, with the smaller basins and higher treatment efficacy of Membrane Bio-Reactors (MBRs) the thermal load from the biology of these systems is approximately 50% of total heat gain.
Relevance
Permit requirements across the United States are likely to become stricter as climate change impacts increase. It is likely that cooling will have to be evaluated on more projects and that plant operations will need a means of better predicting their effluent temperatures. The approach of this model increases the accessibility and transparency of these types of thermal calculations and when shared with the broader water community it is likely that this model will be expanded upon or used with additional software packages (other than excel) that are better suited to process the large amounts of information present.