Long Night of Research

Experience research up close

Visit the two stations of the SCCH

How to predict the condition of a machine?

Forward-looking maintenance using intelligent data analysis is becoming increasingly important. The networking of machines, products and components and other systems involved in the production process is the hallmark of the Internet of Things. Precise prediction of the optimum maintenance time strengthens the company's competitive position.  

Huge data streams (e. g. machine data, process data, etc.) from a wide variety of heterogeneous data sources have to be linked and analysed with each other in order to provide a meaningful basis for decision-making and recommendations for human action. SCCH presents its methods for the realization of predictive analytics and predictive maintenance. Through the use of data mining and machine learning methods, error forecasting models are created in order to find an early warning point and thus facilitate predictive maintenance strategies.  The key is seen as the combination of expert knowledge and data-based error prediction models. The spectrum of applications for these methods ranges from the process industry and production to energy management and the manufacture and maintenance of machines and plants.  


How can data glasses be used to optimize production?

Optimum man-machine interaction, for example during welding, can be achieved with the aid of data glasses. Welding tasks that are not automated often place particularly high demands on their execution. Not only because of the geometry and position of the weld seam to be produced, but often also because of the required work steps and execution accuracy.

In small series production this means a lot of information which is necessary for each welding process. Even for welding professionals, this is a great challenge that takes a lot of time and is prone to errors. In a joint project with an industrial partner, new possibilities for interaction were explored. Due to the special conditions during welding and the limited possibility of interaction with the hands shortly before or during welding, the information display and interaction with Augmented Reality (AR) was used to display additional information in the working environment. The availability of relevant data is required for the information on the current work order. This information includes, for example, the preparation (type of sheet metal, distance between sheets, checking the geometry of the weld gap) and the welding process itself (welding layers, welding material, welding direction and cooling phase). The information is superimposed on the workpieces and optically linked. This minimizes errors. All parameters can be set at the power source via AR glasses. The passage to the device is no longer necessary.

back