孙宗光, 倪一清, 高赞明, 丁皓江. 基于斜拉索振动测量与神经网络技术的斜拉桥损伤位置识别方法[J]. 工程力学, 2003, 20(3): 26-30.
引用本文: 孙宗光, 倪一清, 高赞明, 丁皓江. 基于斜拉索振动测量与神经网络技术的斜拉桥损伤位置识别方法[J]. 工程力学, 2003, 20(3): 26-30.
Z. G. Sun, Y. Q. Ni, J. M. Ko, H. J. Ding. DAMAGE LOCATING OF CABLE-STAYED BRIDGES BASED ON DYNAMIC MEASUREMENT AND NEURAL NETWORK TECHNIQUE[J]. Engineering Mechanics, 2003, 20(3): 26-30.
Citation: Z. G. Sun, Y. Q. Ni, J. M. Ko, H. J. Ding. DAMAGE LOCATING OF CABLE-STAYED BRIDGES BASED ON DYNAMIC MEASUREMENT AND NEURAL NETWORK TECHNIQUE[J]. Engineering Mechanics, 2003, 20(3): 26-30.

基于斜拉索振动测量与神经网络技术的斜拉桥损伤位置识别方法

DAMAGE LOCATING OF CABLE-STAYED BRIDGES BASED ON DYNAMIC MEASUREMENT AND NEURAL NETWORK TECHNIQUE

  • 摘要: 以香港汲水门大桥为背景,探讨了斜拉索张力指标和神经网络技术相结合对桥面结构损伤的定位或分类方法。基于高精度三维有限元模型,采用BP网络,模拟了12种可能的损伤情况的定位。以不同损伤程度下斜拉索张力指标作为神经网络的训练与测试输入,由神经网络的输出来指示损伤位置。该方法的突出优点是只用到少量斜拉索的局部模态的基频,就能获得较好的识别结果。而对少量斜拉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。另外,该方法可十分方便地推广应用于悬索桥的桥面损伤定位。因此,该方法具有重要的实用价值。

     

    Abstract: Taking cable-stayed Kap Shui Mun Bridge as an example, this paper presents a new damage locating method for cable-stayed bridges based on the combination of cable tension index and neural network technique. By using BP network, damage localization for 12 potential damage cases are simulated based on a high-precision FE model. Taking cable tension indices as inputs of neural network for both training and testing, damage locations are indicated by the outputs of the network. The outstanding feature of the method is that a good result can be obtained by using only fundamental natural frequencies of a few stayed cables. Because measurement of fundamental natural frequencies of a few stayed cables is much easier than some other damage-oriented measurement, the method has great practical value. The method can be easily used for damage locating of cable suspension bridges.

     

/

返回文章
返回