What Students Say:
“We approached SWAN because we were in need of case studies for a journal we are compiling as part of our master studies. SWAN immediately offered to connect us with people and through their connections we were able to make an appointment at Vitens, a leading company for drinking water in the Netherlands.”
– Sue van Geijn, Graduate Student at the University of Twente, Netherlands
RiSWP Spotlight: Sanni Eerikäinen, Project Manager, Finnish Water Forum
A Master’s Candidate at Aalto University in Finland, Sanni shared her thesis on Wastewater Treatment Plants and answered questions about the process. View here full thesis HERE and feel free to contact her at: email@example.com
What inspired you to study data analytic solutions related to Finnish municipal wastewater treatment plants?
While studying wastewater treatment, I became interested in process control. It seemed to me that while new analytics tools are quickly adopted for process control in other industries, wastewater treatment plants have used the same tools and strategies for decades – despite the vast research in the field. Seeing this gap between research and practice, I wanted to find out what solutions are actually needed at the plants and what’s hindering the development.
What were the biggest hurdles you faced and how did you overcome them?
As a part of the study, I interviewed over 30 professionals in the field. From all data collection methods, interviews are probably the most vulnerable method for researcher biases and suggestions. Different interviewees also shared some rather contradictory information. Instead of trying to eliminate the subjective views in the data, I decided to do a thematic analysis to highlight them instead. Many findings reflect the interviewee opinions and not verified facts: whether true or false, it’s these opinions that steer the pace and direction of development.
What were your research key takeaways and how can these be applied to international utilities?
- Maintenance of instrumentation in WWTPs is a laborious task and easily disregarded. This often leads to unreliable data quality, which then complicates using the data for control or analysis. When purchasing instrumentation for WWTPs, focus should be shifted from initial price to life cycle costs instead. This would enable technology providers to offer better quality sensors with smaller maintenance need. Alternative way to tackle this issue may be to outsource the whole sensor acquisition, maintenance and data collection, and pay only for the verified data instead.
- Operational reliability won’t be risked for the sake of process optimisation. And it shouldn’t be, since large safety margins are necessary to handle the range of risk scenarios arising from the uncontrollable upstream. Development and adoption of data analytics tools should be first focused on quality assurance of instrumentation, equipment and controllers instead. Once the quality assurance is in place, also process optimisation is possible without the risk of disturbing the process.
- Current research in the field is focused on studying advanced control applications with process models. Though, in the actual process control system there’s a range of limitations that can easily deteriorate the control performance. These include sensor faults, actuator controllability, maintenance and update needs, to name a few. Higher emphasis on control robustness already in the research phase would speed up the adoption to practice.
What advice do you have for other students interested in pursuing research on data-driven topics in water or wastewater?
Data-driven tools are changing the way the world works and they do have the potential to renew also our industry. While deep-diving to this intriguing world, try not to grow too fond of the methods, but focus on the ultimate goal – the results. More than often, the most brilliant idea is the most simple idea.