Link To Latest Report : Final Report.
Project Statement :
Bridges, being critical infrastructure, need consistent monitoring as they are susceptible to damage from environmental factors, aging, and other causes. To address these challenges, advancements in structural health monitoring are being made, integrating new techniques to protect and enhance the longevity of bridges.
SHM has transitioned from manual inspections to automated monitoring, aiding in maintenance of infrastructure and enhancing disaster prevention through effective monitoring. This section explores the evolution of SHM, from traditional methods to sensor-based evaluation. This change shows how important technology is becoming in maintaining structural health. However, regular SHM methods can be expensive and require extensive work and time with sensor-based data collection. The use of technology further signifies a transition towards the proper utilization of data and incorporating advanced techniques such as machine learning to ensure the resiliency and robustness of structures, advancing SHM.
Objectives :
This report explores the use of advanced machine learning techniques, specifically convolutional neural networks (CNNs) and gated recurrent units (GRUs), within the domain of SHM for bridges. The primary aim is not only to explore but also to integrate these state-of-the-art technologies to advance the field forward. The primary objectives of this report are: (1) leveraging advanced machine learning models to improve the identification of structural damage in bridges; (2) enhancing damage predictions, especially during seismic events; and (3) strengthening the overall resilience of bridges. By focusing on these objectives, this study aims to provide novel approaches to challenges in SHM, providing the opportunity for more reliable and efficient monitoring techniques in civil engineering, and providing useful insights for predicting future issues and planning repairs.
Scope :
The scope of this study spans across four applications, each of which addresses the specific aspects of SHM using machine learning. These applications are designed with the primary goal of making significant contributions to the field of SHM.
Research Team :
Principal Investigator : Dr. Islam Mantawy
Co- Principal Investigator : Ravuri Naga Lakshmi Chittitalli.