基于吊索局部振动与神经网络技术的悬索桥损伤定位  

Damage locating of a suspension bridge based on hanger local vibration and neural network technique

作　　者：伍雪南[1] 孙宗光[1] 毕波[2] 苏建[1] 

机构地区：[1]大连海事大学,大连116026 [2]鞍钢附企设计研究院,鞍山114000

出　　处：《振动与冲击》2009年第10期203-206,共4页Journal of Vibration and Shock

基　　金：教育部留学回国科技基金(JYB-04-1)

摘　　要：基于由吊索局部振动测试构建的吊索张力指标,应用神经网络技术对悬索桥结构损伤位置和程度识别进行了探讨。首先应用经校正的三维有限元模型,对该方法进行了较为详细的数值模拟分析。模拟了多种可能的损伤工况。采用BP网络,以不同损伤程度下基于吊索频率计算的张力指标作为网络的训练与测试输入,由网络的输出向量来指示损伤位置及程度。然后,利用面向健康诊断专门设计制作的悬索桥试验模型,针对个别损伤工况进行了损伤识别的模型试验研究。数值模拟和试验研究均获得较好的损伤识别效果。该方法的突出优点是只用到少量吊索的局部模态的基频,就能获得较好的识别结果。而对少量吊索的局部模态的基频测量要比其它面向损伤检测的测量容易得多。因此,该方法具有重要的实用价值。Based on neural network technique and hanger local vibration measurement,identification of damage location and its level for a suspension bridges was discussed.First,damage identification was simulated numerically based on a high-precision FE model validated by experiment.Some potential damage cases for the suspension bridge were simulated.A BP network for damage identification was trained and tested by taking hanger tension indices as input.The index was constructed based on the fundamental natural frequency of hanger local vibration mode.The damage locations and level were indicated by the output vector of the network.Then,a model test for identification of some damage cases was carried out on the test model of the suspension bridge.The model was specially designed aiming at damage identification and health diagnosis of the suspension bridge.The both results of numerical simulation and model test showed the method was effective and practical.The outstanding feature of the method was that a good result can be obtained by using only fundamental natural frequencies of a few hangers;because measurement of fundamental natural frequencies of a few hangers is much easier than some other damage-oriented measurement,the method has great practical value.

关 键 词：悬索桥 损伤识别 吊索局部振动 神经网络 

分 类 号：U448.25[建筑科学—桥梁与隧道工程]