【作者】 殷玉平； 刘剑； 周彤； 丁勇； 李浩恒；

【Author】 Yin Yuping;Liu Jian;Zhou Tong;Ding Yong;LiHaoheng;School of Civil Engineering,Harbin Institute of Technology;Key Lab of Structures Dynamic Behavior and Control(Harbin Institute of Technology),Ministry of Education;Zhong Dian Jian Ji Jiao Expressway Investment Development Company Limited;

【机构】 哈尔滨工业大学； 结构工程与控制教育部重点实验室哈尔滨工业大学； 中电建冀交高速公路投资发展有限公司；

【摘要】 在自然环境因素、长期交通荷载共同作用下,结构不可避免会出现一定程度的损伤,导致承载能力降低,因此需要对桥梁进行监测,根据结构响应输出信号来识别结构模态参数或者物理参数,评估桥梁实际性能,确保结构的可靠性。对于大跨桥梁结构,很难通过主动施加外部激励的方法对桥梁进行激振从而进行参数识别,而且现场采集结构响应信号会受到环境等噪声干扰,给识别效果带来影响。本文将提出一种融合移动荷载作用下桥梁结构反应测试信息的桥梁损伤识别方法。在桥梁运营状态下,桥梁加速度、应变信息较易获得并进行长期记录。本文通过无迹卡尔曼滤波方法对测试信息进行融合作为观测量,并对桥梁损伤参数进行递推估计。通过OpenSees-UKF仿真平台对不同程度预应力混凝土桥梁损伤识别进行了数值仿真分析,结果表明融合信息较单一测试信息可有效提高参数识别精度。更多还原

【Abstract】 Subjected to the environment factors and long-term traffic load,the structure will inevitably be damaged to a certain extent,resulting in the reduction of bearing capacity.Therefore,it is necessary to monitor the bridge,identify the structural modal parameters or physical parameters according to the structural response,evaluate the actual performance of the bridge,and ensure the reliability of the structure.For long-span bridge structure,it is difficult to identify the parameters of the bridge by actively applying the external excitation,and the on-site acquisition of structural response signals will be interfered by the environment and other noise,which will affect the identification result.In this paper,a bridge damage identification method based on the information of bridge structural response under moving load is proposed.Under the condition of bridge operation,the acceleration and strain information of bridge is easy to obtain and recorded for a long time.In this paper,the method of Unscented Kalman filter is used to fuse the test information as the observation measurement,and the damage parameters of the bridge are estimated recursively.Through the OpenSees-UKF simulation platform,the damage identification simulation studies of prestressed concrete bridge are conducted.The results show that the fusion information can effectively improve the accuracy of parameter identification compared with single test information.更多还原