复合材料板变形分布式光纤感知与控制方法研究

发布时间：2018-07-20 11:26

【摘要】：光纤光栅传感器具有体积小、抗电磁干扰、灵敏度高、易于构建分布式传感网络并适合与复合材料结构集成等独特优点,非常应用于航空航天领域。现代飞行器向着多功能、高机动、高可靠等方向发展,具有自诊断、自修复、自适应功能的智能化变体飞行器备受重视。因此,本文提出研究基于神经网络PID算法的类似于机翼柔性后缘的复合材料层合板弯曲变形分布式光纤感知与控制方法,主要工作包括以下几个方面:首先,深入研究了预埋光纤光栅传感器的复合材料层合板制备方法,如烘箱固化与热压罐固化工艺、光纤传感器铺设与引出方式等。为考察固化工艺对于光纤光栅传感器感知性能的影响,研究了复合材料层合板不同固化阶段光纤光栅传感器反射光谱与中心波长偏移量变化特征,为后继固化工艺优化和传感器性能评估提供有益参考。其次,构建了基于光纤光栅传感器的单边固支环氧树脂板与碳纤维复合材料层合板弯曲变形监测系统。借助试验标定方法,得到典型加载条件下板结构弯曲变形与光纤光栅传感阵列中心波长偏移量之间映射关系,从而为板结构变形快速辨识提供依据。再次,针对机翼结构变形控制需求,在传统PID算法基础上,分别给出了基于Hebb单神经元、BP神经网络以及RBF神经网络控制下的PID算法。数值仿真表明,神经网络PID控制算法在变形控制的超调量、调节时间等性能方面均有较大改善。最后,构建了基于BP神经网络PID算法的复合材料层合板变形分布式光纤监测与控制系统,编写了基于Labview的复合材料层合板弯曲变形感知与控制软件,实现对板结构弯曲变形状态的控制。
[Abstract]:Fiber Bragg grating (FBG) sensors have many unique advantages such as small size, anti-electromagnetic interference, high sensitivity, easy to build distributed sensor networks and suitable for integration with composite structures, and are widely used in the field of aeronautics and astronautics. Modern aircraft is developing towards multi-function, high maneuverability and high reliability. Intelligent aircraft with self-diagnosis, self-repair and adaptive function has been paid more attention to. Therefore, this paper presents a distributed fiber sensing and control method for bending deformation of composite laminates, which is similar to the wing flexible trailing edge based on neural network pid algorithm. The main work includes the following aspects: first of all, The fabrication methods of composite laminates for embedded fiber Bragg grating sensors, such as curing process of oven and hot pressing tank, laying and extraction of fiber optic sensors, etc. In order to investigate the effect of curing process on the sensing performance of fiber Bragg grating sensors, the variation characteristics of reflectance spectrum and center wavelength offset of fiber Bragg grating sensors in different curing stages of composite laminates were studied. It provides a useful reference for the optimization of the subsequent curing process and the performance evaluation of the sensor. Secondly, based on fiber Bragg grating sensor, the bending deformation monitoring system of single side clamped epoxy resin plate and carbon fiber composite laminated plate is constructed. With the help of the experimental calibration method, the mapping relationship between the bending deformation of the plate structure and the central wavelength offset of the fiber grating sensor array under typical loading conditions is obtained, which provides the basis for the rapid identification of the plate structure deformation. Thirdly, according to the requirements of wing structure deformation control, the pid algorithm based on Hebb single neuron BP neural network and RBF neural network control is presented based on the traditional pid algorithm. Numerical simulation shows that the neural network pid control algorithm can improve the performance of deformation control such as overshoot and adjusting time. Finally, the distributed optical fiber monitoring and control system of composite laminates deformation based on BP neural network pid algorithm is constructed, and the software of sensing and controlling the bending deformation of composite laminates based on LabVIEW is developed. The control of bending deformation state of plate structure is realized.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TB33;TP212

【参考文献】