WCW26

Coagulation chemistry is a complex process impacted by several variables and is regarded as “the single most important factor affecting treatment plant performance” (AWWA M37, 2011). Changes in source water characteristics due to seasonality, runoff, and other conditions require appropriate adjustments to coagulant and polymer doses to maintain efficient treatment. Adjustment can be challenging, as the full impacts of coagulant dosing decisions are not seen for hours, and in some cases, days, and may be impacted by combinations of several parameters. Predictive models can help operational staff optimize coagulant and polymer dosing decisions during both rapidly changing and stable raw water conditions.


Coagtech has been working with several large drinking water treatment plants with varying source waters, coagulants, clarification technologies, and levels of instrumentation to assess and deploy machine learning for coagulation optimization. Assessments have been completed at water treatment plant serving a combined population of >4M people.


This presentation will provide an overview of coagulation, machine learning, and the key considerations for water treatment coagulation optimization. Specific challenges related to data gathering, data analysis, and modelling will be highlighted. Technology implementation approach, including the central role of the operator, and challenges and benefits of the technology adoption will be shared.