The Institute of Flight Systems and Automatic Control (FSR) at TU Darmstadt has an ongoing research history in the field of predictive maintenance, also referred to as Prognostics and Health Management (PHM). With the aim of predicting a remaining useful lifetime (RUL) of an individual unit, this novel technique fosters new possibilities for cost savings (e.g. reduced downtimes) and maintenance optimization (e.g. spare part logistics). In parallel, it can be used to improve the overall system safety by monitoring the current health status of safety critical components.
In several projects, the FSR has worked on the development of (mainly) data-based diagnosis and prognosis algorithms as well as on concepts for the implementation and realization of condition-based maintenance (CBM) applications. By operating own test stands, the FSR is able to generate realistic degradation data, which is used to support the evaluation and benchmarking of the developed methods. In this context, the FSR is active in all six steps of the OSA-CBM process, reaching from the data acquisition and diagnosis to the prognosis and decision-making. With the use of cost-benefit analyses, also the economic view is accounted for.
A deeper insight into our research is presented in our recent article about predictive maintenance in aviation, published in the “Ingeneurspiegel” (issue March 2017), which can be found here.
Our current projects in this field include:
INDI (“Intelligent Data Utilization in Maintenance“) is a LuFo (Aviation research program of the Federal Ministry of Economics and Energy) funded joint research project with project partner Lufthansa Technik. Within this project, the FSR assesses the development of intelligent algorithms and data analysis strategies in the field of machine learning and artificial intelligence in order to optimize the maintenance of aircrafts. More …
With the SiFliegeR project the FSR investigates how PHM can be leveraged to generate a dynamic and system specific safety analysis, based on the current system health status. The applicability of this method is demonstrated on an unmanned aerial vehicle’s control surface actuation system. More…