Mechanical State Assessment of In-Service Cable-Stayed Bridge Using a Two-Phase Model Updating Technology and Periodic Field Measurements

Abstract

An innovative two-phase model updating technique is developed to accurately assess the mechanical state of an in-service cable-stayed bridge. In the first stage, a sensitivity analysis is performed to select the updating parameters. Objective functions are established according to the measured values and numerical results of the static performance of the in-service cable-stayed bridge. The finite-element model updating of the bridge is developed into a multiobjective optimization problem. The Pareto optimal solution set for the updating parameters is solved using the genetic algorithm. In the second stage, the fuzzy outranking method is adopted to rank the noninferior solutions from the first stage and identify the best compromise solution for the updating parameters. The proposed two-phase model updating technique is applied to assess the dead load mechanical state of Haihe Bridge (old) in Tianjin China. Periodic field measurements are also conducted to detect its actual mechanical state after 10 and 12 years of service. The relative differences between the original numerical results and the periodic measurements of cable forces are up to −17.15%. However, after model updating, the relative differences between the calculated and periodically measured values of cable forces are controlled within ±10% throughout. The results indicate that the updated model can reproduce the periodic field measurements with considerably improved accuracy. Based on the updated model, the vertical deformation of the main span girder is evident, and that of the side span girder is small. The maximum girder vertical deformation of Haihe Bridge (old) is −0.271 m and is located at approximately 3/4 of the main span.