Hybrid simulation of continuous-discrete systems

Abstract

Many process models in the chemical industry involve discrete phenomena superimposed on continuous system dynamics. While the continuous behavior is described by mass, energy and momentum balances, discrete behavior may occur due to physico-chemical discontinuities or discrete logical controller actions. Such systems with interaction between the continuous and discrete system dynamics are classified as Hybrid Systems . For these systems, standard numerical treatment via “continuous” integration methods break down. Equally, discrete simulation e.g. finite state machine cannot handle the continuous dynamics. Hence, a new mathematical framework capable of handling both continuous and discrete aspects of hybrid systems is needed. This paper presents a new environment for modeling and simulation of hybrid systems. It offers a high-livel design language for the automatic or semi-automatic generation of the process models. The efficacy of the simulation language is demonstrated by means of continous-discrete implementation of a Dynamic Matrix Control strategy.