Multivariable computer control of a butane hydrogenolysis reactor: Part I. State space reactor modeling

Abstract A state space model is developed for an existing nonadiabatic packed‐bed reactor which experiences axial and radial concentration and temperature gradients resulting from highly exothermic reactions involving several chemical species. The set of partial differential equations is reduced from three to two dimensions using orthogonal collocation. Further use of orthogonal collocation on the quasi steady state version of the differential equations results in a set of coupled algebraic and differential equations from which a linearized low‐order state space model is obtained. The model parameters are estimated from reactor data, and the fitted model is used for design and implementation of a multivariable stochastic control schemes on the reactor. These applications are described in parts II and III of this paper.