IMMEDIATE SEISMIC RELIABILITY INVESTIGATION FOR TRANSPORT SYSTEM USING DEEP LEARNING

Transport systems such as roads, bridges, and tunnels are critical infrastructures that are vulnerable to earthquakes. In the event of a seismic event, these structures may suffer significant damage that could impact the safety of commuters and the economy. Thus, there is a need for an immediate seismic reliability investigation for transport systems to ensure the safety and resilience of critical infrastructure. This research paper proposes a deep learning-based approach for immediate seismic reliability investigation of transport systems. The approach involves utilizing data from generic structure for k-terminal reliability evaluation sensor networks and developing a DL model for identify the damage state for transport systems. It is crucial to apply precise and effective methods to assess system dependability against probability occurrences in order to optimise mitigation, readiness, reaction, and recovery strategies for the maintenance of infrastructure. The method that is most frequently used to calculate the effect of natural catastrophes on systems of infrastructure still has a significant computing cost, for complex systems. Using a case study of a transportation system, deep learning techniques are used to accelerate seismic reliability assessments. Different deep learning surrogates are built and researched. a complete stand-in for modules like connection analysis, connectivity determination, and connectivity averaging. The efficiency for suggested reliability analysis while obtaining accuracy is demonstrated by resultsfrom a k-terminal study of the transportation system in response to a probable seismic incident. The results demonstrate the potential of deep learning for rapid and accurate seismic reliability investigation of transport systems.