Doctor of Philosophy (PhD)
Crystallization is the main physical separation process in many chemical industries. It is an old unit operation which can separate solids of high purity from liquids, and is widely applied in the production of food, pharmaceuticals, and fine chemicals. While industries in crystallization operation quite rely on rule-of-thumb techniques to fulfill their requirement, the move towards a scientific- and technological- based approach is becoming more important as it provides a mechanism for driving crystallization processes optimally and in more depth without increasing costs. Optimal operation of industrial crystallizers is a prerequisite these days for achieving the stringent requirements of the consumer-driven manufacturing. To achieve this, a generic and flexible model centric framework is developed for the advanced operation of crystallization processes. The framework deploys the modern software environment combined with the design of a state-of-the-art 1-L crystallization laboratory facility. The emphasis is on developing an economically and practically feasible implementation which can be applied for the optimal operation of various crystallization systems by pharmaceutical industries. The key developments in the framework have occurred in three broad categories: i. Modeling: Using an advanced modeling tool is intended for accurate representation of the behavior of the physical system. This is the cornerstone of any simulation, optimization or model-based control approach. ii. Monitoring: Applying a novel image-based technique for online characterization of the particulate processes. This is a promising method for direct tracking of particle size and size distribution with high adaptability for real-time application iii. Control: Proposing numerous model-based strategies for advanced control of the crystallization system. These strategies enable us to investigate the role of model complexity on real-time control design. Furthermore, the effect of model imperfections, process uncertainty and decision variables on optimal operation of the process can be evaluated. Overall, results from this work presents a robust platform for further research in the area of crystal engineering. Most of the developments described will pave the way for future set of activities being targeted towards extending and adapting advanced modeling, monitoring and control concepts for different crystallization processes.
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Ghadipasha, Navid, "A Model-Centric Framework for Advanced Operation of Crystallization Processes" (2017). LSU Doctoral Dissertations. 4304.