Oliveira Cabral, Thiago2025-08-052025https://hdl.handle.net/2097/45215Integrated process systems transform chemical manufacturing by enabling modular, decentralized, and multifunctional operations. The increasing integration of renewable energy technologies and sustainable practices reshapes industrial paradigms, introducing complex dynamics that demand advanced modeling and control strategies. This dissertation presents a robust computational framework grounded in fundamental chemical engineering principles—transport phenomena, reaction kinetics, and thermodynamics—to enable predictive modeling, process design, and decision support in chemical-energy systems, focusing on hydrogen and ammonia production and utilization. Mechanistic models were developed for various applications, including ammonia synthesis and separation, solar-thermal energy integration, carbon vapor deposition, and multiphase fluidization. These models provide a reliable basis for capturing system behavior and guiding operational strategies. Where necessary and computationally tractable, multiscale representations were incorporated to reflect the hierarchical nature of physical phenomena across molecular, mesoscopic, and macroscopic domains. While multiscale effects are inherent in chemical processes, their use in modeling and optimization has historically been limited by measurement and computational challenges. At the smallest scales, atomic and molecular interactions determine surface kinetics and thermodynamic properties. At intermediate scales, transport mechanisms interact with reaction dynamics to shape local process behavior. At the plant level, interconnections among unit operations influence overall performance and economic outcomes. Capturing these interactions is essential for developing accurate and effective models that support informed decision-making. Model predictive control (MPC) was integrated with the developed process models to enable optimal system-level operation. As an optimization-based control strategy, MPC offers a structured approach for handling nonlinear dynamics, constraints, and uncertainty. This work conceptualizes MPC as a supervisory decision-making layer responsible for coordinating energy and material flows across integrated networks. Beyond regulation, MPC is utilized as a planning and scheduling tool for real-time operation under diverse scenarios. While the present framework demonstrates its effectiveness, future research opportunities remain in model reduction, data-driven modeling, and scalable distributed control.en-USHigh-fidelity modelingMultiscale modelingChemical processesDecision-making methodsDissertation