Project Grant 2543135

Federal Project Grant Award Summary

Oregon State University received a $550,000 NSF Engineering (CFDA 47.041) CAREER award effective September 15, 2026, through August 31, 2031, to develop mechanism-driven machine learning (ML) frameworks for water and wastewater treatment systems. The project delivers an application-driven artificial intelligence (AI) framework that integrates operational and water quality data from multiple treatment facilities with hybrid modeling approaches combining physics-based process models and ML components. The research employs mechanistic ordinary differential equation models coupled with ML, physics-informed neural networks, and embedded neural differential equations to model high-risk unit processes, specifically water and wastewater disinfection. Deliverables address critical environmental dataset challenges including data sparsity, autocorrelation, measurement uncertainty, and site-specific variability through time-aware validation, uncertainty quantification, and risk-based performance metrics.

Beyond the technical research outputs, the project generates workforce development products by integrating data science into environmental engineering curricula to address the national need for engineers capable of using AI responsibly. Findings and AI models are designed for dissemination to water utilities and environmental engineers across multiple treatment systems, particularly supporting communities with limited technical resources. The research advances scientific discovery by providing mechanistic insights into how and why treatment processes function at full-scale, ultimately improving the reliability and safety of water treatment infrastructure while reducing operational risk across engineered environmental systems.