极端台风下特大悬索桥的GNSS实时振动监测及动力响应分析

GNSS-based Real-time Vibration Monitoring and Dynamic Response Analysis of a Long-span Suspension Bridge During an Extreme Typhoon Event

特大桥梁结构健康监测系统为桥梁模态识别、损伤探测和安全评估提供了重要参数信息和决策依据。鉴于传统加速度计、计算机视觉、光纤光栅等方法存在位移测量不精确、易受外界环境干扰、仅获取局部变量等问题,构建了全球导航卫星系统(Global Navigation Satellite System, GNSS)特大悬索桥实时振动监测和改进经验小波变换(Empirical Wavelet Transform, EWT)的结构动力响应分析方法。依托由GNSS、加速度计和风速仪等传感器组建的苏格兰福斯公路桥GeoSHM变形监测系统,自2014年起实现对桥塔、主梁的高精度、全自动、全天候、三维实时监测;建立基于改进协方差自回归功率谱的频带划分方法和Pearson相关系数有效固有模态函数筛选准则,提出改进EWT算法进行降噪滤波和位移分解,实现风致桥梁振动响应的动力响应分析。采用上述方法对2020年2月的席亚拉(Ciara)台风GNSS监测数据进行分析,识别出索塔纵向振动位移最大振幅为7.91 cm,动态位移最大振幅为2.84 cm, 2座桥塔上GNSS接收机识别出的模态频率基本吻合,风速与桥塔水平位移呈正相关,索塔的水平位移沿风向增加明显;主梁纵向、横向和竖向振动位移最大振幅分别为5.51、152.47、68.17 cm,其中动态位移最大振幅为3.73、31.93、54.22 cm,与加速度计识别的动态位移及振动频率结果吻合,风速与主梁三维位移亦呈正相关,台风引起的横向振动最大,因此横向位移随风速增加的趋势显著。研究结果表明:构建的GNSS监测技术和改进EWT数据处理方法,可实现特大桥梁风致振动实时监测和动力响应分析,具有重要的科学研究和工程应用价值。

The structural health monitoring system for long-span bridges provides important parameter information and a decision basis for bridge modal identification,damage detection,and safety assessments.Traditional methods such as accelerometers,computer vision,and fiber-optic sensors suffer from various issues such as inaccurate displacement measurements,susceptibility to external conditions,and limited capture of local variables.To address these challenges,the authors constructed a new method for monitoring the real-time vibration of large suspension bridges using the Global Navigation Satellite System(GNSS)and improving the empirical wavelet transform(EWT)for structural dynamic response analyses.The application of the deformation monitoring system GeoSHM for the Forth Road Bridge in Scotland based on the use of GNSS receivers,accelerometers,and wind speed sensors since 2014,realized high-precision,fully automatic,all-weather,and three-dimensional real-time monitoring of the bridge towers and main beams.The frequency band division method-based on the improved covariance autoregressive power spectrum-and the effective intrinsic mode function screening criterion of the Pearson correlation coefficient were established,and the improved EWT algorithm was proposed to perform noise reduction filtering and displacement decomposition to realize dynamic response analyses of the vibration responses of wind-induced bridges.Analysis of GNSS monitoring data during storm Ciara in February 2020 revealed that the maximum amplitude of the longitudinal vibration displacement of the cable tower was 7.91 cm,and the dynamic displacement was 2.84 cm.The modal frequencies identified by GNSS receivers on the two bridge towers were in good agreement,and there was a positive correlation between wind speed and the tower's horizontal displacement.The horizontal displacement of the cable tower increased considerably along the wind direction.The maximum amplitudes of the three-dimensional(longitudinal,lateral,and vertical)vibration displacement of the main beam were 5.51 cm,152.47 cm,and 68.17 cm,respectively.The maximum amplitudes of the dynamic displacement were 3.73 cm,31.93 cm,and 54.22 cm;these values are consistent with the dynamic displacement and vibration frequency results identified by the accelerometer.The wind speed is also positively correlated with the three-dimensional displacement of the main beam.The lateral,typhoon-induced vibration was the largest and the trend of lateral displacement increased as a function of wind speed.The research results demonstrate that the constructed GNSS monitoring technology and improved EWT data processing method can achieve real-time monitoring and dynamic response analysis of wind-induced vibrations in super-large bridge structures with significant scientific and engineering implications.