Multi-Agent Systems for Model-based Fault-Diagnosis in Technical Processes

This research area deals with the design and implementation of a real-time multi-agent system for model-based fault-diagnosis in complex modular production systems. The architecture comprises several hierarchically organized process agents, a simulation agent and a user agent. Key components in model-based diagnosis are simulators. A distributed petri-net simulator and a massive parallel continuous simulator , both based on multiple TMS320C40 signal processors, are used as the core of the simulation agent. System characteristics are continuous operation, handling of uncertain data, focussing of resources on critical events and guaranteeing of response times.