Condition monitoring and predictive maintenance for higher machine availability
Wear and tear on robot gearboxes can lead to unplanned production interruptions, which cost a lot of time and money. The research project "Artificial intelligence for predicting the operational safety and service life of industrial robots" (KIVI), funded by the Bavarian State Ministry of Economics, Energy and Technology (StMWi), is developing an artificial intelligence (AI) toolbox that can be used to predict the service life of individual robot components. The goal is a continuous condition monitoring and predictive maintenance of industrial robots. In this way, maintenance work can be optimally integrated into the production process and maintenance costs can be reduced.
Once data on the operating vibration behavior of the robot components has been obtained using suitable sensors, various artificial intelligence methods analyze this data in order to recognize patterns in the course of wear conditions and learn behavior models from them.
The results of the project - in the form of a prototype AI toolbox - will be integrated into existing production systems for evaluation purposes. A later commercial exploitation has several advantages: It increases plant availability, makes the production process more efficient and conserves resources.