Predictive maintenance reduces unplanned downtime and enables on-demand machine maintenance.
We implement this on the basis of state-of-the-art algorithms from digital signal processing and machine learning. Your user know-how and innovative methods in the field of machine learning allow us to use your machine data to predict when these machines or their components will fail. This enables you to provide predictive maintenance for your systems and minimize unplanned downtimes and collateral damage.
To ensure the predictive maintenance of bearings, gearboxes or pumps, vibration or structure-borne noise is measured using optimized high-performance sensory systems. However, machines also provide valuable information in the normal log files such as power consumption, set parameters, material consumption and completed process steps.
Adaptive solutions for complete sensor systems make it possible to retrofit your systems for predictive maintenance applications.
Our sensor fusion and artificial intelligence research groups are responsible for writing the algorithms for digital signal processing and machine learning. In particular, new algorithms in the field of deep learning open up new possibilities for predictive maintenance. If required, we can also implement energy-autonomous embedded solutions on which the algorithms run and also develop the sensors that provide the required measurement data.