My research interests lie at the intersection of process systems engineering and sustainable chemical processes, with a strong focus on optimizing emerging technologies to enhance energy efficiency and environmental sustainability. The growing global demand for sustainable chemical manufacturing requires innovative approaches to reduce energy consumption, minimize environmental impact, and promote circular economy principles. I apply machine learning tools to address limited data challenges, develop data-driven models, and solve superstructure optimization to identify the optimal technology selection.
Abstract
Machine learning (ML) has emerged as a transformative tool in chemical engineering, offering powerful capabilities for predictive modeling, optimization, and the development of hybrid computational frameworks that extend beyond traditional mechanistic approaches. My doctoral research leverages ML to address complex chemical process challenges through three interconnected case studies. Each case integrates ML with fundamental engineering principles to optimize experimental processes, enhance sustainability, and facilitate scalable industrial applications. The first case explores membrane-based nutrient recovery from wastewater, using ML to model structure-property-process relationships for selective ion separation and membrane design. The second investigates microwave-assisted ammonia synthesis, where ML enables accurate modeling and optimization of reaction conditions with minimal experimental data. The third focuses on pulsed microwave heating for co-production of ammonia and ethylene, applying ML to dynamically control and intensify non-equilibrium reaction systems.
These studies form the foundation of a comprehensive research plan aimed at advancing sustainable chemical process design. In nutrient recovery, my work develops ML-driven optimization of polyelectrolyte-coated nanofiltration membranes and explores superstructure configurations using mixed-integer nonlinear programming. For microwave-assisted ammonia synthesis, I am building a superstructure framework that integrates reactor and process components, assessing trade-offs in efficiency, emissions, and cost. In the third aim, pulsed microwave heating is paired with reinforcement learning to precisely control reaction dynamics and maximize dual-product yields. This work not only contributes to sustainable process development but also demonstrates how ML can enable the practical implementation of advanced chemical technologies with limited data availability.