Design and implementation of a hybrid modelling and simulation strategy for integrated control

The design and implementation of a hybrid modelling and simulation strategy is described. The strategy is based on the use of equations, sequential function charts (Grafcets), rules, and object-orientation. This results in a hybrid model base which supports both the hierarchical representation of system structure and the description of the continuous, discrete, and heuristic components of a model. Coupled with the hybrid model base is the notion of hybrid simulation, based on the concurrent execution of a continuous simulator, a discrete simulator, and a heuristic simulator (inference engine). The three simulators exchange information through interactions which are modelled as notional objects with attributes which specify the nature of the interaction and the types of variables involved in the interaction. The strategy is implemented in, and is fully integrated with, the G2 realtime expert system. This enables the framework to serve a dual purpose as a facilitator of the integration of simulation models with large-scale hybrid systems and a design vehicle for such functions as scheduling, supervisory control, sequence control and production optimisation. The strategy is applied to the modelling and simulation of a rolling mill, to demonstrate its potential.