Methodologies for the optimisation, control and consideration of uncertainty of reactive distillation

The work presented in this thesis is motivated by the current obstacles hindering the implementation of reactive distillation in industry, mainly related to the complexities of its design and control, as well as the impact of uncertainties thereupon. This work presents a rigorous methodology for the optimal design and control under uncertainty of reactive distillation. The methodology can also be used to identify and investigate mitigation strategies for process failures arising due to design and/or operation deficiencies under changed processing conditions, based on the evaluation of different design and/or control alternatives. The first step of the methodology is the simultaneous (MINLP) optimisation of the design and operation of a reactive distillation process superstructure, used to explore the possible steady-state design alternatives available, including ancillary equipment such as pre- and side-reactors, side-strippers and additional distillation columns, based on product-related constraints and a detailed objective cost function. The next step is the investigation of the dynamic control performance of this optimal system, where conventional and advanced process control strategies are considered in order to investigate how robust the system is towards operational disturbances, or whether revising the optimal steady-state design is required. As the optimisation depends heavily on accurate data for reaction kinetics and separation performance, the final step of the methodology is the evaluation of the impact of parameter uncertainty on the performance of the optimal controlled system, including redesigning the controlled system if required. The methodology is demonstrated using a number of industrially relevant case studies with different reaction and separation characteristics in order to investigate how these determine the design and control of an economically attractive and rigorous reactive distillation process. It is demonstrated that the process characteristics have a significant impact on the design of the system, and that auxiliary equipment may be required to meet production specifications and/or to ensure robust controlled behaviour. It is also shown that, under parameter uncertainty, an optimal controlled system may nevertheless face performance issues, and revising the design and/or operation of the process may be required in order to mitigate such situations