(156b) Synthesis of Networks of Reactors Featuring Multiple Residence Time Distribution

Albassam, A., University of California Los Angeles
Manousiouthakis, V., University of California Los Angeles, Los Angeles

In this work, we present a methodology to synthesize globally minimum volume reactor networks, which are allowed to employ multiple reactor types featuring a number of normalized residence time density (NRTd) functions. The synthesis problem is formulated within the Infinite DimEnsionAl State-space (IDEAS) conceptual framework, which is shown to be applicable to the problem at hand. Within the IDEAS framework, network synthesis is formulated as an infinite dimensional linear optimization problem. The IDEAS conceptual framework is realized through solution of a series of finite dimensional linear programs whose optimum values converge to the infinite program’s infimum. Global optimality is established by demonstrating that the input–output information maps of the reactors with general RTd/RTD models satisfy all properties required for the application of the infinite dimensional state-space (IDEAS) approach to the RTd/RTD reactor network synthesis problem. The resulting IDEAS formulation is shown to possess a number of novel properties, which can be used to facilitate its solution. The power of the proposed methodology is demonstrated on a case study featuring a variety of reactor types, including PFR, CSTR, and SLFR (Segregated Laminar Flow Reactor).