Computational and analog modeling of parabolic transport equations using Cellular Neural Networks

This paper is concerned with the numerical and analog modeling issues for a class of parabolic transport equations of the type reaction-advection-diffusion. The computational procedure relies on the Method of Lines concept and the canonical cell-based neural networks paradigm. It is derived from a recently proposed procedure for solving hyperbolic propagation equations with non-standard boundary conditions. The resulted highly structured and repetitive computational device allows both computational and analog modeling of the considered distributed parameter system. The procedure is applied to a nonlinear bioreactor distributed parameter model of the type reaction-advection-diffusion with Hadane kinetic function.