In the context of Industrie 4.0, modular and skill-based production enables the manufacturing of small batch sizes. This leads to increased complexity of dynamic manufacturing contexts due to more constellations of cyber-physical production modules, skills, and products. However, even existing approaches aiming to enhance efficient and resilient production, fault detection applications, usually requiring big amounts of data or knowledge, remains mainly unnoticed. Against this background, it is the objective of this paper to develop a model that extends skill-based production by means of incorporating automated fault detection. This includes the identification of required fault detection components as well as its integration into the process control for executing skills.
Hence, to achieve this objective, state-of-the-art approaches of fault detection are adapted and integrated into the skill model. Finally, the elaborated model is prototypically implemented in a realistic demonstration environment to come up with a proof of concept. Since it is difficult to directly model faulty behavior of Cyber-Physical Production Modules in flexible production their nominal behavior during the execution of skills is modelled and compared with its actual behavior. Based on deviations and defined thresholds symptoms are generated. The developed components are synchronized with their skill execution.
On Bringing Fault Detection to Skill-Based Production
Lect.Notes Mechanical Engineering
European Symposium on Artificial Intelligence in Manufacturing ; 2023 ; Kaiserslautern, Germany September 19, 2023 - September 19, 2023
2024-06-27
11 pages
Article/Chapter (Book)
Electronic Resource
English
Bringing automation to production
Automotive engineering | 1986
|Skill training for production of memorized movement pattern
Engineering Index Backfile | 1968
|Online Contents | 2008