压电阻抗法层合板界面损伤识别研究

发布时间：2018-08-19 16:10

【摘要】：复合材料层合结构的破坏往往起源于界面层。在长期荷载作用下,大型、复杂的结构,由于自然灾害、外部环境、结构本身荷载等因素都会使界面产生损伤,且这种损伤是不易察觉的,从而使结构的力学性能严重退化,甚至发生整体破坏。为了充分利用层合结构的性能,提高其使用价值,对界面力学行为的研究十分必要。同时,为保证结构的安全性,需采取有效手段对其进行实时健康监测。压电阻抗法(EMI)是一种基于智能材料的结构健康检测技术,因其对结构早期微小损伤比较敏感,常被用在土木工程等领域。通过建立压电阻抗模型,提取包含结构损伤信息的EMI信号并进行分析,从而对损伤进行判定。但在实际工程中,EMI信号数据庞大,怎样对这些数据进行快速有效的处理一直是研究的重点。压缩感知(Compression sensing)理论与BP神经网络相结合为大量数据的处理提供了一种可行的方式。压缩感知方法可以对信号进行无失真压缩,大大降低了它的维数。本文对复合材料层合板的弱界面问题进行研究。引入粘弹性界面本构关系模拟界面特性,利用状态空间法求解弹性力学解析解。同时,考察含压电层层合板的机电效应,导出电导纳关系式,验证EMI技术的可行性。然后,建立各向同性及各向异性层合板EMI有限元模型,通过赋予层间弹簧单元不同的刚度系数模拟结构不同界面损伤程度,提取出相应的EMI信号。利用压缩感知理论对信号进行压缩,以提高信号传输和存储能力。为了进一步降低信号的维数,再进行主成分分析,将分析后的主成分特征值输入BP神经网络,实现损伤信号的鉴定与分类。结果表明EMI技术能够对层合结构界面损伤进行检测,同时压缩感知与BP神经网络相结合的方式能够有效降低数据的维数,实现信号的快速传输与存储,并能对界面损伤进行定量归类,为工程健康监测提供了一种可行的方式。
[Abstract]:The failure of composite laminated structure often originates from interfacial layer. Under the action of long-term load, large and complex structures, due to natural disasters, external environment, load of the structure itself and other factors will cause damage to the interface, and this damage is not easy to detect, thus making the mechanical properties of the structure seriously degraded. There is even total destruction. In order to make full use of the properties of laminated structure and improve its use value, it is necessary to study the mechanical behavior of interface. At the same time, in order to ensure the safety of the structure, it is necessary to take effective means to carry out real-time health monitoring. Piezoelectric impedance method (EMI) is a kind of smart material based structural health detection technology. It is often used in civil engineering because of its sensitivity to early minor damage of structures. The piezoelectric impedance model is established to extract and analyze the EMI signal which contains the damage information of the structure, so as to determine the damage. However, in practical engineering, EMI signal data is huge, how to deal with these data quickly and effectively has been the focus of research. The combination of compressed perceptual (Compression sensing) theory and BP neural network provides a feasible way to deal with a large number of data. The compression sensing method can compress the signal without distortion, which greatly reduces its dimension. In this paper, the weak interface of composite laminates is studied. The viscoelastic interface constitutive relation is introduced to simulate the interface properties and the state space method is used to solve the analytical solution of elastic mechanics. At the same time, the electromechanical effect of piezoelectric laminated plates is investigated, and the admittance relation is derived to verify the feasibility of EMI technology. Then, the EMI finite element model of isotropic and anisotropic laminated plates is established, and the corresponding EMI signals are extracted by simulating the different interface damage degree of the structure with different stiffness coefficients of the interlayer spring element. The compression sensing theory is used to compress the signal to improve the signal transmission and storage capacity. In order to further reduce the dimension of the signal, principal component analysis (PCA) is carried out, and the eigenvalue of the principal component is input into BP neural network to identify and classify the damage signal. The results show that EMI can detect the interface damage of laminated structure, and the combination of compression sensing and BP neural network can effectively reduce the dimension of data and realize the fast transmission and storage of signals. The interface damage can be classified quantitatively, which provides a feasible method for engineering health monitoring.
【学位授予单位】：宁波大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU317

【参考文献】