Publikationen

In prognostics and health management (PHM), the task of constructing comprehensive health indicators (HI)from huge amounts of condition monitoring data plays a crucial role. HIs may influence both the accuracyand reliability of remaining useful life (RUL) prediction, and ultimately the assessment of system’s degradationstatus. Most of the existing methods assume apriori an oversimplified degradation law of the investigatedmachinery, which in practice may not appropriately reflect the reality. Especially for safety–critical engineeredsystems with a high level of complexity that operate under time-varying external conditions, degradationlabels are not available, and hence, supervised approaches are not applicable. To address the above-mentionedchallenges for extrapolating HI values, we propose a novel anticausal-based framework with reduced modelcomplexity, by predicting the cause from the causal models’ effects. Two heuristic methods are presented forinferring the structural causal models. First, the causal driver is identified from complexity estimate of thetime series, and second, the set of the effect measuring parameters is inferred via Granger Causality. Once thecausal models are known, off-line anticausal learning only with few healthy cycles ensures strong generalizationcapabilities that helps obtaining robust online predictions of HIs. We validate and compare our framework onthe NASA’s N-CMAPSS dataset with real-world operating conditions as recorded on board of a commercial jet,which are utilized to further enhance the CMAPSS simulation model. The proposed framework with anticausallearning outperforms existing deep learning architectures by reducing the average root-mean-square error(RMSE) across all investigated units by nearly 65%.

Visualizations have a distinctive advantage when dealing with the information overload problem: since theyare grounded in basic visual cognition, many people understand them. However, creating the appropriaterepresentation requires specific expertise of the domain and underlying data. Our quest in this paper is tostudy methods to suggest appropriate visualizations autonomously. To be appropriate, a visualization hasto follow studied guidelines to find and distinguish patterns visually, and encode data therein. Thus, a visu-alization tells a story of the underlying data; yet, to be appropriate, it has to clearly represent those aspectsof the data the viewer is interested in. Which aspects of a visualization are important to the viewer? Canwe capture and use those aspects to recommend visualizations? This paper investigates strategies to recom-mend visualizations considering different aspects of user preferences. A multi-dimensional scale is used toestimate aspects of quality for charts for collaborative filtering. Alternatively, tag vectors describing chartsare used to recommend potentially interesting charts based on content. Finally, a hybrid approach combinesinformation on what a chart is about (tags) and how good it is (ratings). We present the design principlesbehindVizRec, our visual recommender. We describe its architecture, the data acquisition approach with acrowd sourced study, and the analysis of strategies for visualization recommendation