Home ▻ Research sites▻Ward & Burke Centre▻Machine Learning for Wastewater Treatment Plant Operation

Machine Learning for Wastewater Treatment Plant Operation

Machine learning for noisy missing and uncertain data

PROJECT OVERVIEW

The WWTP-ML project develops machine learning methods to improve the operation of wastewater treatment plants through better use of operational data. It focuses on building robust data pipelines and AI models that can handle noisy, incomplete, and uncertain sensor data commonly found in real plant environments. The research supports fault detection, predictive maintenance, and data-driven decision-making, while also exploring approaches for real-time and near real-time process monitoring and control.

WHY IT MATTERS

  • Wastewater treatment plants generate large volumes of complex, imperfect data that is often underutilised
  • Faults in sensors or processes can go undetected, affecting compliance and efficiency
  • Improving data quality and interpretation enables more reliable plant operation
  • Predictive maintenance helps reduce unexpected failures and operational downtime
  • AI-based tools support better decision-making for plant operators and engineers
  • Enables a shift toward more automated, resilient wastewater treatment systems
  • Provides a foundation for scalable digital infrastructure for future smart water systems

ABSTRACT

Machine learning is increasingly being applied to wastewater treatment to improve monitoring, control, and operational efficiency. However, real-world plant data is often noisy, incomplete, and uncertain, limiting the effectiveness of traditional analytical approaches.

The WWTP-ML project, funded by Ward and Burke Construction Ltd., Insight Research Ireland Research Centre, and the Ward and Burke Centre for Infrastructure Research and Innovation, addresses these challenges by developing robust machine learning methods for wastewater treatment plant operation. The project is co-led by Prof. Eoghan Clifford (Civil Engineering) and Dr James McDermott (Computer Science), University of Galway.

The research focuses on three key areas: (i) developing automated data ingestion and preprocessing methods capable of handling missing and inconsistent sensor data; (ii) implementing anomaly detection and predictive maintenance techniques to identify faults in both plant processes and sensing equipment; and (iii) exploring real-time and near real-time modelling approaches for improved forecasting, simulation, and operational control.

By integrating data engineering, machine learning, and domain-specific process understanding, the project aims to enable more reliable, efficient, and intelligent operation of wastewater treatment systems, supporting the transition toward data-driven infrastructure management.

Principal Investigators:

Dr Eoghan Clifford, Ph.D, Civil Engineering, University of Galway 

Dr James McDermott, B.Sc., PhD, Computer Science, University of Galway

James

James McDermott |   Senior Lecturer and Director of Research & Graduate Studies

James McDermott is Senior Lecturer and Director of Research & Graduate Studies in the School of Computer Science, University of Galway, Ireland. He has previously worked and studied in Hewlett-Packard, University of Limerick, University College Dublin, and Massachusetts Institute of Technology. His research interests are in artificial intelligence, including genetic programming, evolutionary optimisation, and deep learning, with applications in sustainability and AI music. He has chaired international conferences including EuroGP, EvoMUSART, and GECCO GECH Track, and is a member of the Genetic Programming and Evolvable Machines journal editorial board, and associate editor of the ACM SIGEvolution newsletter.
in Connect with James