基于最敏感斜拉索张力指标的斜拉桥主梁损伤识别方法

发布时间：2019-05-10 20:57

【摘要】：斜拉桥结构主梁的损伤会引起恒载内力重分布,进而引起斜拉索张力变化。根据这一规律,结合BP神经网络算法,研究并提出基于最敏感斜拉索张力指标的斜拉桥主梁损伤识别方法。通过分析斜拉索张力对主梁损伤的敏感性,确定对主梁损伤最敏感的斜拉索,并将该斜拉索在主梁损伤前后的张力变化率定义为最敏感斜拉索张力指标;将主梁发生不同位置、不同程度损伤时对应的各最敏感斜拉索张力指标作为BP神经网络的输入并训练网络,根据训练误差自动确定网络的隐含层结构,由训练好的BP神经网络的输出确定主梁的损伤位置及损伤程度。以实验室独塔斜拉桥试验模型为例,采用有限元软件ANSYS建立其空间板壳有限元模型,运用给出的主梁损伤识别方法对该斜拉桥试验模型的主梁损伤进行识别,结果表明:该方法仅用1组4根斜拉索便可以较好的识别率实现对斜拉桥主梁不同位置、不同程度单点损伤的识别。
[Abstract]:The damage of the main girder of cable-stayed bridge will cause the redistribution of internal force under dead load and the change of tension of cable-stayed cable. According to this rule, combined with the BP neural network algorithm, the damage identification method of the main beam of cable-stayed bridge based on the most sensitive cable tension index is studied and proposed. By analyzing the sensitivity of cable-stayed cable tension to the damage of main beam, the cable-stayed cable which is the most sensitive to main beam damage is determined, and the tension change rate of the cable-stayed cable before and after damage is defined as the most sensitive tension index of cable-stayed cable. The tension index of each of the most sensitive cable-stayed cables with different positions and different degrees of damage is taken as the input and training network of BP neural network, and the hidden layer structure of the network is determined automatically according to the training error. The damage location and damage degree of the main beam are determined by the output of the trained BP neural network. Taking the test model of single tower cable-stayed bridge in laboratory as an example, the finite element model of space plate and shell is established by using the finite element software ANSYS, and the damage identification method of the main beam is used to identify the damage of the main beam of the test model of the cable-stayed bridge. The results show that only one group of 4 cable-stayed cables can be used to identify the single point damage of the main beam of cable-stayed bridge in different positions and degrees.
【作者单位】：石家庄铁道大学工程力学系;
【基金】：国家自然科学基金资助项目(50778116) 河北省自然科学基金资助项目(E2012210061) 河北省教育厅重点项目(ZH2012068) 河北省科学技术研究与发展计划项目(11215611D) 河北省人力资源和社会保障厅项目(436018)
【分类号】：U448.27

【参考文献】