Introducing Explicit Causality in Object-oriented Hybrid System Modeling
Résumé
Along with the rapid development of embedded devices and network technology, the area of Cyber
Physical Systems (CPS), has arisen. In terms of modeling and simulation, CPS—like many technical systems—have a
hybrid nature, i.e., discrete-event behavior and continuous-time dynamics have to be integrated with each other.
Basically, this integration is supported by modern object-oriented modeling paradigms such as Modelica®. The
equation-based concept resolves the causality between interconnected components, which qualifies this modeling
scheme for complex multi-domain systems. However, in hybrid systems, explicit causality is required to correctly
manage iterative events. This paper highlights these issues, including algorithmic loops and instantaneous multiple
updates, which essentially arise from incompatibilities between the object-oriented concept and specific discrete-event
phenomena. We discuss several possible solutions and introduce the concept of re-allocating the objects’ behavioral
intelligence.
Domaines
Modélisation et simulation
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...