Structural Health Monitoring in Smart Cities: An Artificial Intelligence Approach to Infrastructure Resilience

Abstract

Structural Health Monitoring (SHM) is vital for the protection of urban buildings as more cities become smart. This work suggested an AI-based SHM framework to support the enhanced resilience, safety and upkeep of critical infrastructure. WSNs, edge computing and a deep learning model mixing CNN and LSTM units are part of the framework. As a result, architects can track any obvious concerns in structure in near real-time with vibration and strain sensors placed on both urban bridges and high-rise buildings. The model uses data collected in simulations and the real world to discover errors in structures, ensuring an accuracy of over 96%. Moreover, when edge-based processing is used, both latency and bandwidth needs are minimized, making the system capable of handling many large-scale deployments. A case study of a bridge in Bangalore indicates that the approach can monitor continuously, detect faults early and warn in advance. The new process is shown to have a 35% lower rate of false positives than theoretical threshold methods. The results emphasize that AI-led SHM systems play a key role in predictive maintenance and strengthening urban infrastructure in these rapidly changing smart urban environments.