Robust nonlinear control using bilinear matrix inequalities with application to a batch crystallization process

While linear robust control theory is relatively well understood with many standard techniques for control system analysis and controller design, methods for designing robust nonlinear controllers are underdeveloped. This constitutes a significant gap in chemical process control theory, as real chemical processes have significant nonlinearities associated with their startup, shutdown, and normal operation.; The main objective of the present study is to develop a rigorous approach to design robust nonlinear controllers for chemical processes. The proposed design methodology integrates robust control theory results with nonlinearity inversion techniques. The proposed methodology is applied in a batch crystallization process. An identification and optimization procedure for a pharmaceutical, batch crystallization process is also presented, for complete understanding of the modeling issues entailed.; The controller design problem necessitates the solution of an optimization problem under bilinear matrix inequality constraints, which is a nonconvex optimization problem, has been shown to be NP-hard, and its efficient solution is also an open research problem. Various solution strategies have been incorporated to a branch and bound algorithm solving the aforementioned optimization problem, and were compared for various controller designs.