Major Research Aspects

It is the key objective of SelSus to build an approach for systematic knowledge generation in the design phase and knowledge gathering and refining in the usage phase of the assembly station. This knowledge can be integrated in an assembly system and the knowledge transfer can be offered as added value to the end-user of the machine.

The main ideas of the SelSus project is to develop a new life cycle based and distributed probabilistic diagnostic and predictive maintenance environment for repair and renovation activities which is fully aware of the condition and history of the machine components within highly complex manufacturing facilities. The system integrator delivers his knowledge embedded into the machine by providing a data model, an algorithmic prediction model as well as degradation and deterioration models as an added value. The models actively assist the maintenance personnel in planning, diagnosing and selecting the best strategy for renovation and repair of equipment/components. The models refine the knowledge during the usage phase by constant communication with the maintenance and repair personnel.

Therefore the development of the following objectives are focused in SelSus:

 

Self-diagnosing, self-healing devices

SelSus will develop self-healing devices with enhanced capabilities on condition monitoring, sensing and self-adaption, with each device having significant capabilities beyond that of an ordinary device. 

Sensor Networks

A sensor network that consists of a variety of sensor nodes will be developed. Their capability for distributed storage and analysis, interoperable and delay-tolerant communication, will pave the way for a truly scalable network of sensors which will support adaptable plug-and-produce assembly stations. 

Decision Models

A decision model on how to act when either unforeseen malfunctions or disruptions arise or the long-term prediction models advise to renovate or substitute specific subsystems.

Integrative Model for Predictive Maintenance

For complex interdependent manufacturing systems the only way to simultaneously minimize maintenance and repair costs, probability of failure as well as energy consumption is to perform on-going life-long assessment of self-healing of equipment and continuous prediction of future performance trends and possibilities of potential malfunctions through forecasting the degradation and deterioration trends of equipment/components.

Machine Learning and Data Mining

Aiming to ensure easy model construction and less reliance on expert model specification, SelSus will provide novel machine learning and data mining techniques to enable the construction of dynamically updating, self-learning prediction models.

Methodologies to build up Self-Diagnosis and Predictive Models

By providing compact methodologies, SelSus will enable design engineers to build up self-diagnosis and predictive models with reasonable demands regarding time consumption as well as complexity.