基于振型叠加法的柔性结构形态重构研究

发布时间：2018-06-06 00:32

本文选题：有限元分析 + 形态重构　；参考：《上海大学》2016年硕士论文

【摘要】：大型柔性结构广泛应用于航空航天领域,例如太阳能帆板、大飞机机翼和卫星板状天线等。这些航天结构部件一般具有刚度小、质量轻和面积大等特点,在长期运行时需要保证一定的运行精度,但是在太空中该类结构易受到轻微扰动而发生结构变形和低频振动,这对航天器造成了极大的安全隐患。因此,研究柔性结构的形态重构对航天部件的健康监测具有非常重要的意义。针对柔性结构的形态重构,一般是在柔性结构表面布设传感器测量结构形态变化的物理信息如应变、加速度等,然后利用形态重构算法建立一种从测量的物理信息到全局位移场的转换关系,从而实现结构的形态重构。本文以航天结构健康监测为研究背景,针对板状柔性结构模型,将基于光纤布拉格光栅(FBG,Fiber Bragg Grating)传感器的应变测量网络、基于振型叠加法的位移估算方法和计算机图形处理技术结合起来进行柔性结构形态重构的方法技术研究。首先,本文针对所研究的柔性结构进行了基于有限元方法的结构力学分析。在ANSYS有限元分析软件中创建了目标结构的有限元模型并进行了模态分析,得到了结构的应变振型和位移振型。通过静力分析和谐响应分析对结构的变形形态做了分析并了解其结构形变特征。同时结构静动态力学分析获得的结构变形响应如应变和位移为结构形态重构数值仿真提供了对比分析数据。其次,对柔性结构形态重构进行了理论推导和数值仿真。振型叠加法首先利用少量测量应变信息和应变振型叠加关系获取振型坐标,然后利用振型坐标对位移振型进行线性组合求出结构的全局位移,从而建立了一种从测量离散应变信息到结构全局位移场的转换关系。本文采用C#语言结合ANSYS二次开发接口技术对柔性结构的形态重构算法进行了数值仿真,仿真结果表明针对柔性结构简单变形和复杂变形,此重构技术具有良好的形态重构效果。同时,对感知网络传感器布局进行了优化研究并对重要参数进行了分析。本文选用结构形态重构误差作为传感器布局优化准则,利用模拟退火优化算法对传感器的位置进行了优化研究,传感器位置通过优化后明显提高了结构形态重构的重构效果;然后分别分析了传感器数量和方向对结构形态重构效果的影响;最后对传感器的位置和方向做了多目标优化分析,结果表明传感器多目标优化的结构形态重构效果更加显著。此外,本文对激励频率、激励幅值、模态阶数和条件数的作用进行了参数研究。最后,搭建了实验平台并对形态重构方法进行了实验验证和分析。在实验中,首先在柔性结构表面布设FBG应变传感器获取结构的离散应变信息,然后利用基于振型叠加法的位移重构算法对研究对象进行重构并通过可视化平台进行实时显示,从而达到对柔性结构形态变化的实时监测。实验中分析了静态变形和动态变形两种情况,为了评估实验结果,激光位移传感器在特定点测得的位移被用作参考位移,实验结果表明,在静态和动态实验中,重构位移和测量位移具有较好的吻合性,验证了基于振型叠加法的结构位移重构技术的有效性。
[Abstract]:Large flexible structures are widely used in the field of Aeronautics and Astronautics, such as solar panels, large aircraft wings and satellite plate antennas. These spaceflight components generally have the characteristics of small stiffness, light mass and large area, so it is necessary to ensure a certain operating precision in the long run, but this kind of structure is easily disturbed in space. Structural deformation and low frequency vibration have caused great potential safety hazards to the spacecraft. Therefore, it is of great significance to study the shape reconstruction of flexible structures for the health monitoring of space components. Strain, acceleration and so on, and then use the morphological reconstruction algorithm to establish a transformation relationship from the physical information of the measurement to the global displacement field, so as to realize the structure reconfiguration. This paper takes the space structure health monitoring as the research background, and aims at the plate flexible structure model, based on the fiber Prague grating (FBG, Fiber Bragg Grating) sensor. The strain measurement network, based on the displacement estimation method of the mode superposition method and the computer graphics processing technology, combines the method and technology of the configuration reconstruction of the flexible structure. Firstly, this paper analyzes the structural mechanics analysis based on the finite element method for the flexible structure studied, and creates the mesh in the ANSYS finite element analysis software. The finite element model of the standard structure is carried out and the modal analysis is carried out. The strain and displacement modes of the structure are obtained. Through the analysis of the static analysis and the harmonic response analysis, the deformation characteristics of the structure are analyzed and the structural deformation characteristics are understood. At the same time, the structural deformation response, such as strain and displacement, is the structural shape weight of the structure. The numerical simulation provides the comparative analysis data. Secondly, the theoretical derivation and numerical simulation of the shape reconstruction of the flexible structure are carried out. The vibration mode superposition method first uses a small amount of measurement strain information and the strain mode superposition relation to obtain the vibration shape coordinates, and then uses the mode coordinate to calculate the global displacement of the structure by linear combination of the displacement modes. A transformation relationship from measuring discrete strain information to structural global displacement field is established. This paper uses C# language and ANSYS two development interface technology to simulate the morphological reconstruction algorithm of flexible structure. The simulation results show that the reconstruction technique has a good shape for the simple deformation and complex deformation of the flexible structure. At the same time, the layout of sensor network sensor is optimized and the important parameters are analyzed. In this paper, the structure shape reconstruction error is selected as the optimization criterion of sensor layout, and the location of the sensor is optimized by simulated annealing optimization algorithm. The structure of sensor is improved obviously after the optimization of the sensor position. The effect of the reconfiguration of the morphologic reconfiguration is analyzed, and the influence of the number and direction of the sensor on the structure reconfiguration effect is analyzed. Finally, the multi-objective optimization analysis is made on the position and direction of the sensor. The results show that the structural shape reconstruction effect of the multi-objective optimization of the sensor is more significant. In addition, the excitation frequency, excitation amplitude, and mode are also analyzed in this paper. In the experiment, the FBG strain sensor is set up on the flexible structure surface to obtain the discrete strain information of the structure, and then the displacement reconstruction algorithm based on the mode superposition method is used to study the research object. In the experiment, two cases of static and dynamic deformation are analyzed in the experiment. In order to evaluate the experimental results, the displacement of the laser displacement sensor at a particular point is used as the reference displacement. The experimental results show that the static and dynamic results are in the static and dynamic conditions. In the state experiment, the reconstructed displacement and the measured displacement have a good agreement, which verifies the effectiveness of the structural displacement reconstruction technology based on the mode superposition method.
【学位授予单位】：上海大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：V414;TP212.9

【参考文献】