Discover the Editor's Choice: From each issue of the leading journal Water Science & Technology the core team of editors select one outstanding paper to share across IWA Publishing platforms.
IWA Publishing is pleased to announce that the 35th Editor's Choice Paper, chosen by Daniele Cecconet is now Open Access:
A. Bernardelli, S. Marsili-Libelli, A. Manzini, S. Stancari, G. Tardini, D. Montanari, G. Anceschi, P. Gelli and S. Venier
Why was this paper selected for Editor's Choice?
Wastewater treatment plants (WWTPs) are complex systems in which biological, chemical and physical processes are carried out to achieve contaminants removal from aqueous waste streams. In addition, nutrients can be recovered to further valorize wastewater. Nevertheless, WWTPs are high energy intensive, and proper control and operation is vital to reduce their energy impact. In order to achieve this goal, several plant-wide models have been proposed in the last 15 years.
Bernardelli et al. propose a novel and innovative approach for the predictive control of a WWTP based on machine learning and in particular on neuro-fuzzy computing. A model predictive controller was designed and tested onsite in a large-scale WWTP (about 500.000 P.E.), and allowed for the detection of total nitrogen peaks sufficiently in advance to adapt the air supply. This translated into a reduction of the overall energy consumption of the WWTP while still achieving high nutrients removal.
Although nowadays machine learning has become a trending topic, very few studies show a practical application on full scale WWTPs with reliable and consistent results. This study allows a deep understanding of the application of machine learning to WWTP control and optimization while targeting also a non-expert audience. I appreciated reading a paper that represents scientific innovation for a topic of practical relevance and hope that this also applies to the readership of this issue of Water Science & Technology.
Daniele Cecconet, Editor.