[Back]

B. Heinzl, W. Kastner:
"Towards Model-driven Development of Hybrid Simulation Models in Industrial Engineering";
Talk: 44th Annual Conference of the IEEE Industrial Electronics Society (IECON2018), Washington D.C., USA; 2018-10-21 - 2018-10-23; in: "Proceedings of the 44th Annual Conference of the IEEE Industrial Electronics Society", IEEE, (2018), 3588 - 3593.

In order to analyze and optimize energy efficiency of complex industrial facilities, methods for hybrid discrete/continuous modeling and simulation are necessary that allow to incorporate thermal-physical dynamic behavior as well as discrete aspects like entity flow into an integrated simulation model while at the same time remaining manageable for large-scale applications. One promising approach follows component-based hierarchical modeling based on a formal system specification, called hyPDEVS. However, this approach faces challenges for widespread adoption in industry due to the generic nature of the formalism and related difficulties in applying it to real-world scenarios. In an attempt to support the development of hybrid simulation models based on hyPDEVS, we present a domain-specific model abstraction for industrial facilities that is intended to allow developing application models in a platform-independent manner. Thereby, we follow a model-driven engineering paradigm that promotes the separation of model specification from the concrete implementation, i.e. source code. Compared to typical co-simulation solutions, this approach provides better semantic integration of discrete and continuous modeling aspects, thereby promoting modularity and reuse of model components.

Biological system modeling;Ovens;Computational modeling;Context modeling;Production;Analytical models;Adaptation models;Model-driven Engineering;PIM;Hybrid Simulation;DEVS;Cube;Metamodel

http://dx.doi.org/10.1109/IECON.2018.8592811

Project Head Wolfgang Kastner:
ASPeCT