SUPERvisory control system for plant-wide OPTImization of wastewater treatment plant operation - SUPEROPTI - Department of Systems and Control

SUPERvisory control system for plant-wide OPTImization of wastewater treatment plant operation - SUPEROPTI

Principal investigator:

Duration:

October 1, 2021 – September 30, 2024

Funding:

ARRS - Slovenian Research Agency, L2-3166, co-financing from
Public Utility JP Centralna čistilna naprava Domžale-Kamnik d.o.o.,
Public Utility Komunala Kranj, javno podjetje d.o.o.,
Public Utility Komunala Novo mesto d.o.o.,
Kolektor Sisteh d.o.o.

Abstract:

Wastewater treatment plants (WWTPs) are public facilities that purify urban and industrial wastewater before its discharge into the environment. Beyond their primary role, WWTPs are increasingly recognised as valuable sources of energy (e.g., chemical energy in wastewater organic fractions) and resource recovery (e.g., phosphorous and nitrogen). Emerging challenges include energy-positive sewage treatment, wastewater reuse, nutrient recovery, and carbon neutrality.

As WWTPs expand their functionalities, they have become complex and demanding to operate. They consist of complex biological processes and other process units that interact with each other via recycling loops. Moreover, they aim to achieve various operational goals that often conflict with one another. Therefore, their management requires a plant-wide perspective, with multi-objective optimisation at the highest level.

In this project, a supervisory control system for plant-wide optimisation was designed. The approach focused on optimising effluent quality (EQI) and operational cost (OCI) indices at the highest operational level. The methodological framework was based on reduced models of performance indices and simplified predictive control. The developed approach was validated through simulation using the WWTP benchmark simulation model and performing dynamic optimisation of the oxygen set-point concentration in an aerobic reactor. The original contributions of the project are a plant-wide perspective and WWTP optimisation at the highest operational level, the selection of key input and optimisation variables using data-driven random forest models, enhanced explainability and interpretability of model input variables by incorporating SHAP values, the identification of non-linear neuro-fuzzy models for EQI and OCI indices, the development of a predictive algorithm for dynamic optimisation of the indices, analysis of Pareto front optimal solutions and evaluation of long-term plant operation. The optimisation was supported by research on lower-lever control methods, which provide the foundation for supervisory control.

A key contribution is also the design, testing and implementation of advanced control and optimisation solutions in real wastewater treatment plants. The performance of the WWTPs was evaluated using process and energy benchmarking methods. Several advanced solutions were tested or implemented at full-scale plants, including energy cost optimisation by balancing on-site electricity generation with plant demand through dynamic control of biogas cogeneration units, advanced feedforward-feedback control of ferric chloride dosing for efficient phosphorus removal, optimisation of control and operational parameters of the biological stage to improve treatment performance and reduce energy and chemical consumption of the WWTP. The implemented solutions contributed to lower operational costs and improved effluent quality of wastewater treatment plants.

Project partners:

Selected publications:

VREČKO, Darko, HVALA, Nadja, BABIČ, Rok. Design and implementation of the SRT control at the Ljubljana WWTP. Water science & technology. [Online ed.]. Sep. 2025, vol. 92, iss. 5, str. 720-731, ilustr. ISSN 1996-9732. https://iwaponline.com/wst/article/92/5/720/109236/Design-and-implementation-of-the-SRT-control-at, DOI: 10.2166/wst.2025.126. [COBISS.SI-ID 249466115].

HVALA, Nadja, VREČKO, Darko, CERAR, Peter, ŽEFRAN, Gregor, LEVSTEK, Meta, VRANČIĆ, Damir. Energy cost optimisation in a wastewater treatment plant by balancing on-site electricity generation with plant demand. Water. Apr. 2025, vol. 17, iss. 8, [article no.] 1170, str. 1-17, ilustr. ISSN 2073-4441. https://www.mdpi.com/2073-4441/17/8/1170, DOI: 10.3390/w17081170. [COBISS.SI-ID 232805379].

HVALA, Nadja, KOCIJAN, Juš. Input variable selection using machine learning and global sensitivity methods for the control of sludge bulking in a wastewater treatment plant. Computers & chemical engineering. [Print ed.]. 2021, vol. 154, str. 107493-1-107493-10. ISSN 0098-1354. https://doi.org/10.1016/j.compchemeng.2021.107493, DOI: 10.1016/j.compchemeng.2021.107493. [COBISS.SI-ID 74445571].

VRANČIĆ, Damir, BISTÁK, Pavol, HUBA, Mikuláš, MOURA OLIVEIRA, Paulo. A new closed-loop control paradigm based on process moments. Mathematics. Jan. 2025, vol. 13, iss. 2, [article no.] 244, str. 1-27, ilustr. ISSN 2227-7390. https://www.mdpi.com/2227-7390/13/2/244, DOI: 10.3390/math13020244. [COBISS.SI-ID 222754051].

KARER, Gorazd, VREČKO, Darko, HVALA, Nadja, ŠKRJANC, Igor. Development and implementation of a wastewater-treatement-plant model for optimization purposes. V: 2024 IEEE 18th International Conference on Control & Automation (ICCA) : 18-21 June 2024. Reykjavík, Iceland. Piscataway: IEEE, 2024. Str. 115-120, ilustr. IEEE International Conference on Control and Automation (Online). ISBN 979-8-3503-5440-9. ISSN 1948-3457. https://ieeexplore.ieee.org/document/10591917, DOI: 10.1109/ICCA62789.2024.10591917. [COBISS.SI-ID 204049923].

DOMINKOVIĆ, Lana, HVALA, Nadja, VREČKO, Darko, BOSHKOSKA, Biljana Mileva. Efficiency and explainability in wastewater treatment plant : a machine learning approach to cost management and effluent quality. V: AMCIS 2024 proceedings : elevating life through digital social entrepreneurship. AMCIS 2024, Salt Lake City, Utah, August 15-17, 2024. [S. l.]: Association for Information Systems, 2024. Str. 1-10. https://aisel.aisnet.org/amcis2024/dsa/dsa/2. [COBISS.SI-ID 206527747].

Department of Systems and Control