论文部分内容阅读
将模态分析与神经网络技术相结合,以某双塔斜拉桥为例,对其拉锁损伤情况进行分析。分析过程中,考虑桥梁拉索结构的单构件损伤、2个构件损伤、3个构件损伤3类损伤工况,采用模态频率指标作为神经网络的输入参数,建立BP神经网络模型。结果表明,基于模态频率和BP神经网络的损伤识别方法可用于桥梁损伤辨别,以及损伤位置及损伤程度识别。
Combining modal analysis and neural network technology, taking a double-tower cable-stayed bridge as an example, the zipper damage situation is analyzed. In the process of analysis, three kinds of damage conditions of the bridge cable structure, such as single-component damage, two-component damage and three-component damage, are considered. The modal frequency index is taken as the input parameter of the neural network to establish the BP neural network model. The results show that the damage identification method based on modal frequency and BP neural network can be used to identify the damage of bridge and identify the damage location and damage degree.