Optimization of retrofit strategies for urban bridge networks before earthquake via dynamic-static integration

Abstract

Addressing earthquake resistance in urban areas, the intervention and retrofit of bridge networks are crucial considerations. While previous studies have predominantly focused on the importance and retrofit aspects of bridges, there has been limited exploration of optimization issues. This study introduces a method for assessing node importance and optimizing retrofit strategies for bridge networks before earthquake. Combining the Personal PageRank (PPR) approach with the technique of ideal solution similarity order preference (TOPSIS) yields static importance evaluation results for nodes, reducing the subjectivity of their weights in multi-metric evaluation works. Employing a sandpile model to simulate post-earthquake traffic congestion, the static results are then refined based on the model. The study establishes a dynamic-static integration approach to assess bridge importance, and genetic algorithms are applied to optimize the retrofit strategy. The practicality of the dynamic and static integration evaluation methods is verified through a numerical example, demonstrating that the optimized strategy effectively improves retrofit performance.