Digital Twin Generalization with Meta and Geometric Deep Learning

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Nov 11, 2024
Raffael Theiler Olga Fink

Abstract

Deep digital twins (DDTs) are deep neural networks that encode the behavior of complex physical systems. DDTs are excellent system representations due to their ability to continuously adapt to operational changes and their capability to capture complex relationships between system components and processes that cannot be explicitly modeled. For this challenge, DDTs benefit greatly from recent success in geometric deep learning (GDL) which allows the integration of information from multiple systems based on schematic representations. A major challenge in training DDTs is their dependence on the quality and representativeness of training data, especially under the dynamic conditions typical in prognostics and health management (PHM). Recent developments in differentiable simulation present new opportunities for optimizing the training data representativeness. In this thesis, we propose a novel meta-learning framework that trains DDTs using the output from differentiable simulators. This setup enables active optimization of training data sampling through gradient computation, enhancing training speed, robustness, and data representativeness. We extend this framework to address challenges in multi-system data integration in power grids and fault detection in railway traction networks. By applying our framework, we aim to tackle significant challenges in forecasting, anomaly detection and sensor-fault analysis using advanced data fusion techniques. Our approach promises substantial improvements in DDT robustness and operational efficiency, with its effectiveness to be demonstrated through empirical studies on both simple and complex case studies within the power systems domain.

How to Cite

Theiler, R., & Fink, O. (2024). Digital Twin Generalization with Meta and Geometric Deep Learning. Annual Conference of the PHM Society, 16(1). https://doi.org/10.36001/phmconf.2024.v16i1.4176
Abstract 98 | PDF Downloads 80

##plugins.themes.bootstrap3.article.details##

Keywords

digital twins, meta-learning, data-fusion, multi-system information fusion, task generalization, anomaly and sensor-fault detection

References
Aivaliotis, P., Georgoulias, K., Arkouli, Z., & Makris, S. (2019, January). Methodology for enabling Digital Twin using advanced physics-based modelling in predictive maintenance. Procedia CIRP, 81, 417–422. doi:10.1016/j.procir.2019.03.072
Antonova, R., Yang, J., Jatavallabhula, K. M., & Bohg, J. (2022, June). Rethinking Optimization with Differentiable Simulation from a Global Perspective (No. arXiv:2207.00167). arXiv.
Biggio, L., Bendinelli, T., Kulkarni, C., & Fink, O. (2022, June). Dynaformer: A Deep Learning Model for Ageing-aware Battery Discharge Prediction (No. arXiv:2206.02555). arXiv.
Booyse, W., Wilke, D. N., & Heyns, S. (2020, June). Deep digital twins for detection, diagnostics and prognostics. Mechanical Systems and Signal Processing, 140, 106612. doi: 10.1016/j.ymssp.2019.106612
Cuadra, L., Salcedo-Sanz, S., Del Ser, J., Jim´enez-Fern´andez, S., & Geem, Z. W. (2015, September). A Critical Review of Robustness in Power Grids Using Complex Networks Concepts. Energies, 8(9), 9211–9265. doi: 10.3390/en8099211
Hospedales, T., Antoniou, A., Micaelli, P., & Storkey, A. (2020, November). Meta-Learning in Neural Networks: A Survey (No. arXiv:2004.05439). arXiv. doi: 10.48550/arXiv.2004.05439
Hu, Y., Miao, X., Si, Y., Pan, E., & Zio, E. (2022, January). Prognostics and health management: A review from the perspectives of design, development and decision. Reliability Engineering & System Safety, 217, 108063. doi:10.1016/j.ress.2021.108063
Jardine, A. K. S., Lin, D., & Banjevic, D. (2006, October). A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mechanical Systems and Signal Processing, 20(7), 1483–1510. doi:10.1016/j.ymssp.2005.09.012
Liao, W., Bak-Jensen, B., Radhakrishna Pillai, J., Wang, Y., & Wang, Y. (2022). A Review of Graph Neural Networks and Their Applications in Power Systems. Journal of Modern Power Systems and Clean Energy, 10(2), 345–360. doi:10.35833/MPCE.2021.000058
Soleimani, M., Campean, F., & Neagu, D. (2021). Diagnostics and prognostics for complex systems: A review of methods and challenges. Quality and Reliability Engineering International, 37(8), 3746–3778. doi: 10.1002/qre.2947
Wang, J., Lan, C., Liu, C., Ouyang, Y., Qin, T., Lu, W., . . . Yu, P. (2022). Generalizing to Unseen Domains: A Survey on Domain Generalization. IEEE Transactions on Knowledge and Data Engineering, 1–1. doi: 10.1109/TKDE.2022.3178128
Zhou, K., Liu, Z., Qiao, Y., Xiang, T., & Loy, C. C. (2022). Domain Generalization: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1–20. doi: 10.1109/TPAMI.2022.3195549
Section
Doctoral Symposium Summaries