Process Intensification of Sustainable Production Systems through Integrated Modeling and Optimization

Abstract

This thesis aims to advance the sustainability of emerging resource production processes through the application of process intensification and integrated modeling and optimization techniques. Two process optimization frameworks are developed and tailored for the optimization of sustainable agriculture and aviation fuel production processes, which are applied through case studies. The first study develops an optimization framework to determine optimal operating strategies in monoculture and polyculture hydroponic systems considering uncertainty and disturbances. A key novelty of this work is the development of a polyculture hydroponic model incorporating interspecies nutrient interactions and dynamic environmental factors into the optimization problem, offering insights for system management and sustainability. Mechanistic nutrient uptake and growth models are validated using experimental data to effectively capture system dynamics, and used to improve resource efficiency while accounting for parameter uncertainty and external disturbances. A case study of hydroponic polyculture soybean and tomato plants demonstrates the benefits of this approach. Results show that hydroponic systems increase yield by over 60% compared to traditional farming. Compared to monoculture hydroponics, polyculture methods reduce nitrogen consumption by 40% and increase annual profit by 3.91% per kilogram of fruit. These findings highlight the importance of nitrogen supply management and demonstrate how computational optimization can advance sustainable agriculture. The second study determines optimal process designs of intensified catalytic distillation (CD) columns for Alcohol-to-Jet (ATJ) sustainable aviation fuel (SAF) production to improve aviation sector sustainability. Sequential CD columns with separate oligomerization and hydrogenation reactions are compared to a fully intensified CD column with simultaneous reactive sections. Although prior studies emphasize the importance of ATJ kinetic modeling, validated models and experimental data remain limited. Accordingly, a first-order kinetic model containing 46 parameters is developed using experimental data and a genetic algorithm to support the SAF CD model. A key contribution is the application of an enhanced parallel hybrid deterministic–stochastic algorithm (eHDSA) for the optimization of SAF CD systems with multiple reactive sections and integer decision variables, including the quantity and location of reactive stages. The eHDSA is modified from prior work to improve performance for complex SAF optimization problems, achieving average total annualized cost (TAC) reductions of 40% compared to a purely stochastic approach. Replacing two sequential CD columns with a single intensified column yields TAC reductions exceeding 86%, with sensitivity analyses indicating strong design robustness. Results demonstrate that CD intensification can reduce capital and energy requirements, improving resource efficiency and environmental sustainability for SAF production within the ATJ pathway.