Search

Industries occasionally discharge slugs of concentrated pollutants to municipal sewers. These industrial discharges can cause challenges at wastewater treatment plants (WWTPs) and reuse systems. For example, elevated total organic carbon that is refractory through biological wastewater treatment increases the required ozone dose, or even exceeds the capacity of the ozone unit, resulting in a treatment pause or diversion. So, alert systems are necessary for potable reuse. Machine learning has many advantages for alert systems compared to the status quo, fixed thresholds on single variables. In this study, industrial discharges were detected using supervised machine learning and hourly data from sensors within a WWTP and downstream advanced treatment facility for aquifer recharge. Thirty-five different types of machine learning models were screened based on how well they detected an industrial discharge using default tuning parameters. Six models were selected for in-depth evaluation based in their training set accuracy, testing set accuracy, or event sensitivity: Boosted Tree, Cost-Sensitive C5.0, Oblique Random Forest with Support Vector Machines, penalized logistic regression, Random Forest Rule-Based Model, and Support Vector Machines with Radial Basis Function Kernel. After optimizing the tuning parameters and variable selection, Boosted Tree had the highest testing set accuracy, 99.2%. Over the 5-day testing set, it had zero false positives and would have detected the industrial discharge in 1 h. However, setting fixed thresholds based on the maximum normal datapoint within the training set resulted in nearly as good testing set accuracy, 98.3%. Overall, this study was a successful desktop proof-of-concept for a machine learning-based alert system for potable reuse.

Membrane bioreactors (MBRs) are increasingly applied to achieve pathogen removal in water recycling schemes. To date, MBRs have been validated in an arbitrary manner, which has led to accreditation varying on a scheme-by-scheme basis and consequently an inequitable treatment of potential health risk. The aim of this thesis is to provide the information required in order to successfully validate a MBR, including: 1) current understanding of the pathogen removal mechanisms, 2) quantitative information on pathogen removal performance and variability, 3) knowledge of how operation and hazardous events influence removal performance and 4) a strategy to assure removal performance continually, through the use of an online monitoring. Surrogate measures for pathogen removal including indicator microorganisms, physiochemical and operational parameters were collected over the course of 18 months during a rigorous sampling campaign of 11 full scale MBRs resulting in a data set with over 250 removal data points for each pathogen group, paired with operating data. Statistical analysis of this data set determined that higher removal was favoured under operation at low range transmembrane pressure, flux and suspended solids concentration and high range pH, temperature, solids retention time and hydraulic retention time. In addition, it was determined that MBRs could consistently achieve 5th percentile log removal values exceeding 1.5, 4.0 and 2.0 for virus, bacteria and protozoa, respectively. High permeate turbidity was shown to correlate with lower removal performance at full scale and a site specific method was proposed and demonstrated at lab scale to calibrate permeate turbidity with removal to allow for continuous monitoring. Hazardous event scenarios, perceived to reduce removal performance, were also investigated. Chemical cleaning was found to have a slightly negative, but recoverable, impact on removal, depending on membrane pore size. Membrane breakage resulted in severely compromised removal, but was recoverable due to activated sludge plugging of defects in the hollow fibre membranes evaluated. Overall, this thesis provided detailed quantitative evidence to support creation of more appropriate and streamlined national validation guidelines for MBRs in Australia and, in doing so, has furthered the knowledge to support health risk management of MBRs internationally.

Membrane bioreactors (MBRs) can provide effective treatment for many wastewater contaminants including chemicals and microorganisms. Operational performances for the removal of these contaminants are most typically characterised under what are considered to be normal operating conditions. However, all MBR systems are continuously subjected to the risk of deviations in operating conditions during what have been termed ‘hazardous events’. Hazardous events may include such things as sudden changes in source water composition, extreme weather events, human error and mechanical malfunctions. Depending on both the likelihood and the consequences of these events, they may ultimately define the treatment reliability and level of risk regarding meeting final water quality objectives. This chapter describes potential hazardous events and their impact on MBR operation.Expected consequences are identified, along with techniques for assessing the likelihoods for some types of hazardous events. Finally, recommendations are made for the management of hazardous events through engineered redundancy and multiple barrier treatment systems.