LAUSR

Abstract In order to analyze and optimize energy efficiency of complex industrial systems, methods for hybrid simulation are necessary that incorporate discrete as well as continuous aspects while at the same remain manageable for large-scale applications. To this end, we compare two approaches that facilitate hybrid simulation: Co-simulation couples multiple simulation environments that form a single model and exchange data at runtime. Typical coupling strategies that introduce communication delays are problematic if discrete variables are involved. The second approach is based on an extended DEVS formalism that facilitates integration of hybrid aspects on the component level and avoids communication delays during simulation. Both approaches are demonstrated on an example of an industrial conveyor oven. DEVS-based hybrid simulation shows to enable better integration of discrete/continuous model aspects also on the semantic level and can potentially achieve higher simulation speed and accuracy.