A Sensor Reduced Machine Learning Approach for Condition-based Energy Monitoring for Machine Tools

With the ongoing digitalization of industrial production, innovative ways of creating energy transparency on the shop floor are emerging. This paper presents a sensor reduced approach to enable condition-based energy monitoring for different degrees of machine data availability. It differentiates between scenarios in which a wide range of machine data can be accessed and thus, machine learning approaches can be applied, and others in which only basic process information can be correlated to data from mobile power measurements. The presented approach is deployed and discussed for an EMAG machine tool in the ETA research factory at the Technische Universität Darmstadt.