21st Presentation | Interpretable Deep Learning for Fatigue damage analysis in primary aerospace structures | Dimitrios Zarouchas
NanoAI welcomes Dimitrios Zarouchas who will present his work, titled “
Interpretable Deep Learning for Fatigue damage analysis in primary aerospace structures”.
As always, an open discussion will follow.
When? January 22nd, 15:30 Greece
Description:
The field of prognostics has gained attention recently in various industries with the aim to optimize maintenance tasks, boost operational efficiency, and prevent costly downtime. Central to prognostics is the Remaining Useful Life (RUL), representing the critical time before system’s failure. Deep learning advancements facilitate RUL forecasting by extracting features from diverse data formats such as time series, images, or sequences thereof, in one, two, or three dimensions, respectively. Yet, predicting RUL from image sequences often relies heavily on resource-intensive techniques. To address challenges with high-dimensional data and unreliable models, this study introduces ISTRUST (Interpretable Spatiotemporal TRansformer for Understanding STructures), an innovative Transformer-based architecture. ISTRUST tackles the dual challenges posed by high-dimensional data and the black-box nature of existing models. Leveraging Transformers’ attention mechanism. Evaluated on fatigue-loaded composite samples showcasing crack propagation, ISTRUST interprets the relation between cracks and RUL via the attention mechanism. The results substantiate its capacity to interpret and clarify instances in which predictions may exhibit variability in accuracy.
Dr. Dimitrios Zarouchas is Associate Professor and the Director of Center of Excellence in AI for Structures, Health Management and Prognostics at the Aerospace Engineering Faculty of Delft University of Technology, the Netherlands. His research interests focus on structural health management and predictive maintenance (Condition-Based Maintenance), damage mechanics and fatigue analysis of lightweight polymer composite structures, structural reliability and stochastic finite element analysis. The main goal of his team is the development and deployment of intelligent cyber-physical structural systems with state awareness capabilities for enabling real-time diagnostics and prognostics. Dr. Zarouchas has published more than 100 peer-reviewed journals and he has been leading multi-million research and innovation projects, funded by EU and National schemes.
Subscribe to our mailing list: https://lists.iit.demokritos.gr/mailman/listinfo/nanoai
You can find more info on this upcoming talk as well as on our previous talks at http://nano-ai.iit.demokritos.gr/