Real-Time Structural Health Monitoring Using Physics-Driven Digital Twins | Shady Adib | DSC MENA 25
Structural Health Monitoring (SHM) is crucial for ensuring the safety and longevity of infrastructure, but traditional methods often struggle with challenges such as data scarcity, computational inefficiency, and the inability to predict rare or unforeseen failure scenarios. In this session, Shady Adib introduces a transformative approach to SHM by integrating digital twins with advanced technologies like physics-informed machine learning, generative AI, and dynamic diffusion models, which offer real-time damage identification and predictive maintenance with remarkable accuracy.
The tutorial focuses on how generative AI addresses the challenge of data scarcity by synthesizing realistic failure data, enabling the development of robust models. Additionally, physics-informed machine learning integrates domain-specific knowledge with data-driven models, ensuring that predictions are both precise and interpretable. The session also explores dynamic diffusion models for probabilistic anomaly detection, which are effective in handling noisy, real-world sensor data.
Shady further demonstrates how the integration of self-healing materials within digital twin frameworks can lead to resilient, adaptive structures capable of autonomous repair.
Through practical case studies using tools like MATLAB, ANSYS, and Python, attendees will gain actionable insights on implementing these methodologies in both research and industrial applications.
This tutorial is aimed at researchers and engineers in civil, mechanical, and aerospace engineering, providing a comprehensive understanding of how next-generation digital twins can revolutionize structural health monitoring and damage detection. Attendees will leave with the knowledge to tackle current challenges and contribute to the development of smarter, more resilient infrastructure.
This tutorial by Shady Adib was held on April 10 at DSC MENA 25 ONLINE.