UC San Diego

Inland waterways infrastructure such as miter gates are subject to damage like cracking and corrosion due to long (~50 years) service lives with extensive water exposure. With the advancement of modern sensing technologies, there’s a vast potential for Structural Health Monitoring (SHM) to transition into a more intelligent and efficient technology that can integrate multiple data sources for enhanced damage diagnostics and inform predictive inspection and maintenance strategies. This research presents a comprehensive optimization framework for the diagnosis and prognosis of such infrastructure. The framework first proposes a novel iterative global-local method for efficient and accurate forward modeling of structural damage in miter gates. It then develops an innovative diagnostic and prognostic framework that not only integrates multiple data sources for structures with multi-failure modes but also analyzes the environmental factors influencing SHM, offering insights into the challenges and solutions for real-world inspections. Furthermore, it introduces a physics-informed inspection planning framework, underpinned by model-based diagnostics and prognostics, leveraging the benefits of digital twin and deep learning technologies. This work represents a significant advancement for a certain class of SHM, providing a robust methodology for improving the lifespan and ensuring the safety of critical waterway infrastructure, marking a crucial step toward the future of infrastructure inspection and maintenance.