Machine Learning based Health Status Assessment of Super-span Suspension Bridges

Abstract

With the rapid development of bridge design and analysis theory, building materials development and construction technology, more and more long-span bridges are built across rivers and seas. At the same time, the bridge construction scheme of over 1000 meters of bridges across rivers and seas is becoming more and more mature. Among them, the mainstream long-span bridges are suspension bridge (SB). The accuracy of research on key structure technology of bridge body needs to be continuously improved to ensure the structural safety of long-span bridges. Therefore, based on machine learning (ML) technology, the health status of super long span SBs is evaluated in this paper. The purpose and significance of long-span bridge structure health monitoring are briefly analyzed. Through the analysis of ML neural network algorithm, the evaluation model index layer is determined; finally, this paper takes the supporting project as an example to evaluate the safety status of the completed SB, which verifies the feasibility and effectiveness of the algorithm in this paper.