Grant: Alliance Grants

Climate change and increasing chemical and energy costs are threatening the integrity of drinking water systems and making water treatment more expensive than ever before. In this context. utilities need new ways of working that minimize costs while still ensuring that users consistently receive clean, safe, and acceptable drinking water. Machine learning (ML), a subset of artificial intelligence (AI), is a powerful tool for understanding and optimizing engineered systems but has made limited inroads into the water industry beyond university research studies. As a large and growing crown corporation, the Ontario Clean Water Agency (OCWA) is in a strong position to explore and capitalize on new technology trends like ML and to pass the benefits along to the utilities and communities that it works with, including many underserved and Indigenous communities in Northern Ontario. We propose to develop a ML framework for optimizing the removal of manganese, a common drinking water contaminant with negative health and aesthetic impacts, at two water treatment plants (WTPs) operated by OCWA with historic manganese exceedances. We will accomplish this by reviewing and assessing existing data at the participating WTPs and developing ML models of compliance and operational resiliency at each plant to provide early warning of compliance and operational issues. We will also develop a ML-based optimization tool that will provide real-time decision-support for WTP operators to help them optimize chemical dosing for manganese removal. Achieving the objectives of this project will allow OCWA to proactively adapt manganese removal to anticipate changing source water quality, ultimately ensuring that these WTPs produce water that is safe and acceptable. This project will train three graduate students over three years. As digital management of water infrastructure increases so will the need for data and AI literacy among water professionals and operational staff. The ML models in this project will be co-developed by YorkU researchers and senior OCWA technical staff and shared with OCWA staff and municipal partners along with YorkU HQP at a tech transfer event.