基于动静载荷数据的桥梁贝叶斯有限元模型多目标更新方法OA

To address the low prediction accuracy and limited engineering applicability of single-objective optimization method in traditional finite element model updating,a multi-objective updating method of Bayesian finite element model based on static and dynamic load data is proposed to improve the simulation ability of the model to the structural response of continuous beam bridge under composite load.The global dynamic characteristic parameters such as the first three frequencies and modes of the structure are obtained by dynamic load test,and the multi-objective non-dominated sorting genetic algorithm II(NSGA-II)is used to synchronously optimize the frequency residual and modal confidence criterion residual,ensuring the accurate calibration of the dynamic characteristics of the finite element model.Based on the deflection data of the key sections obtained from the static load test,a multi-level verification system was established to verify the local response of the modified model.The results show that the proposed multi-objective optimization method can effectively coordinate the conflicts between different objectives,and the response frequency error of the modified model is controlled between 0.70%-4.02%.The static load test shows that the model deflection prediction error range is 2.97%-9.27%,which is at an acceptable level.The research conclusions can provide theoretical and methodological support for health monitoring and performance evaluation of similar bridge structures.