论文部分内容阅读
为提高对悬索桥损伤结构的测量精度,提出将吊索张力指标和神经网络技术相结合对悬索桥结构损伤识别的方法.基于高精度三维有限元模型,模拟7种可能的损伤情况的定位.采用BP神经网络,以不同损伤程度下吊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及程度.该方法的突出优点是只利用少量吊索的局部模态的基频,就可获得较好的识别结果.而对少量吊索局部模态的基频测量要比其他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.
In order to improve the measuring accuracy of the damage structure of the suspension bridge, a method of combining the sling tension index and the neural network technology to identify the damage of the suspension bridge is proposed. Based on the high-precision three-dimensional finite element model, seven possible damage locations are simulated. The neural network is used as training and testing input of neural network under different damage degree, and the position and degree of damage are indicated by the output of neural network.The prominent advantage of this method is that it can utilize only a few slings Frequency, the better recognition result can be obtained, while the fundamental frequency measurement of a small number of slings is much easier than other damage-oriented measurements.Therefore, this method has important practical value.