中文摘要：

基于长期监测数据并采用主成分分析方法建立了润扬大桥悬索桥实测模态频率的环境条件归一化方法,有效地分离了环境因素变化对模态频率的影响,并在此基础上联合应用欧氏距离判别函数和假设检验方法识别模态频率的异常变化.分析表明,该方法可以识别出悬索桥模态频率0.1%的异常变化,具有较强的损伤敏感性,并且不需要监测环境因素变量,适用于悬索桥结构的在线监测与预警.此外,实测分析表明,悬索桥整体状态预警应选择非线性主成分分析方法并采用高阶模态频率作为监测参数.

英文摘要：

Making use of long-term monitoring data of Runyang Suspension Bridge,the environmental-condition-normalized method for measured modal frequencies is formulated based on principal component analysis to effectively eliminate the influences of environmental conditions.The abnormal changes of measured frequencies are further detected by means of the euclidean distance function and hypothesis tests.Analysis results reveal that the proposed method provides a good capability for detecting the damage-induced 0.1% abnormal changes of the measured modal frequencies.And the proposed method needs no monitoring of environmental conditions and hence is suitable for real-time monitoring and warning of suspension bridges.In addition,the measured results indicate that the nonlinear principal component analysis is preferred for condition monitoring by using modal frequencies of higher vibration modes.