Modelling and Control of Cumene Production Process

Abstract

Cumene, a member with substantive consideration in the alkyl aromatic family of newlinehydrocarbons is an inevitable building block in the petrochemical manufacturing newlineindustry. The predominant portion of cumene is used for the production of several newlinechemicals like phenol and acetone which plays multiple role in various chemical and newlineindustrial applications. The unique properties of cumene like high boiling point, newlineflammability and solubility in organic solvents drives its demands in end use industries newlinelike chemical, automotive, paint and plastic. It accelerates the cumene production newlineprocess in the forecast era. Over the last two decades, the cumene production industry newlinehas undergone dramatic changes with the goal of enhancing product quality and lowering newlinealkylation temperature. As one of the largest industries in the world, one way to newlinepotentially meet these performance requirements is to introduce additional design newlineparameters. Hindustan Organic Chemicals Limited (HOCL), Cochin India, has a cumene newlineplant for the production of phenol and acetone. Considering the significance of cumene newlineproduction in the market and the challenges facing in the cumene plant, this research newlineseeks to model the cumene reactor in this plant based on an energy efficient design. newlineIn this dissertation, the modelling of cumene reactor which utilizes Solid newlinePhosphoric Acid (SPA) catalyst is studied. Both the steady state and dynamic models are newlinetypically represented by Partial Differential Equations (PDEs), throwing open a newlinechallenging modelling problem. Orthogonal Collocation Method (OCM) also known as newlinePseudo Spectral Method (PSM) is a numerical strategy for solving partial newlinedifferential equations based on continuing functions or piecewise continuing newlinefunctions. This method has been used for solving steady state and dynamic modelling newlineequations. The collocation approach satisfies the differential equation exactly at a set of newlinecollocation points.