基于图像监控的光纤绕线机排线系统研究

发布时间：2018-05-18 20:41

本文选题：光纤绕线机 + 图像监控　；参考：《北方工业大学》2017年硕士论文

【摘要】：在惯性导航系统中,陀螺仪作为一种新型的敏感元件,为惯性导航系统提供姿态,角速度等重要信息。目前,其被广泛应用于航空航天、航海和国防科工等重要邻域,具有重要的战略意义。针对光纤陀螺仪中重要敏感元件光纤陀螺,在绕制过程中出现的精度差,控制系统落后、绕线效率低、一致性差等问题,本文提出用机器视觉辅助检测改善加工工艺的方法。本文主要内容包括以下几个方面:(1)分析了传统光纤绕线机的机械结构及其相互运动关系。为代替人工拨线,设计了拨线机构。对比了多种光纤绕线工艺的优劣,并对传统光纤绕线机采用的四极对称绕线工艺进行了详细的分析,制定了基于图像监控的绕线工艺流程。(2)完成了光纤绕线机视觉监控系统中摄像机、镜头和光源的选型。分析了多种图像采集系统安装方式的特点,确定明场漫射背光照明是最适合本系统的安装方案。(3)设计了获取光纤拨线轮感兴趣区域的算法,解决了由于摄像机拍摄到图像数据量较大而且含有大量不需要信息的缺点。比较了多种光纤轮廓图像增强、图像分割以及边缘检测算法,最终确定了以中值滤波,动态迭代阈值进行图像分割,然后进行canny边缘检测的基本图像处理过程。(4)设计了从光纤轮廓感兴趣区域获得光纤波峰位置的算法。其中,针对光纤轮廓波峰位置检测的不确定性、间断以及错误波峰位置问题,利用插值和滤波的方法,从而提高寻找波峰位置稳定性。最后根据波峰出现的位置规律,得到了当绕线时出现间隙故障以及重叠故障时的图像处理及监控算法。(5)通过分析(4)中获取的波峰位置信息,设计了预测光纤波峰出现位置的线性拟合预测算法,实验表明该预测算法预测效果良好。实现了通过光纤的预测位置得到拨线轮运动增量,最终完成光纤绕线机的自动拨线控制。(6)设计了基于ARM和FPGA的多轴运动控制器,替代原有的PLC控制方案,提高了控制系统的灵活性。实验表明,该多轴同步控制器具有良好的同步控制效果。
[Abstract]:In inertial navigation system, gyroscope, as a new sensitive element, provides important information for inertial navigation system, such as attitude, angular velocity and so on. At present, it is widely used in aerospace, navigation, defense science and technology and other important neighborhood, has important strategic significance. In order to solve the problems such as poor precision, backward control system, low winding efficiency and poor consistency of fiber optic gyroscope (fog), which is an important sensitive element in fiber optic gyroscope, this paper puts forward a method of improving machining technology by machine vision aided detection. The main contents of this paper are as follows: (1) the mechanical structure of the traditional fiber winding machine and its relationship with each other are analyzed. In order to replace the manual wire dialing mechanism is designed. The advantages and disadvantages of various fiber winding techniques are compared, and the four-pole symmetrical winding technology used in the traditional fiber winding machine is analyzed in detail. The visual monitoring system of optical fiber winding machine is composed of camera, lens and light source. In this paper, the characteristics of various installation methods of image acquisition system are analyzed, and it is determined that the bright field diffuse backlighting is the most suitable installation scheme for this system. The algorithm for obtaining the region of interest of the optical fiber dial wheel is designed. The problem is solved because of the large amount of image data captured by the camera and the large amount of information that is not needed. Several image enhancement, image segmentation and edge detection algorithms are compared. Finally, median filter and dynamic iterative threshold are used to segment the image. Then the basic image processing process of canny edge detection. 4) an algorithm for obtaining the peak position of optical fiber wave from the region of interest of fiber contour is designed. In order to solve the problem of the uncertainty, discontinuity and error peak position of fiber profile wave peak detection, interpolation and filtering methods are used to improve the stability of finding the peak position. Finally, according to the position rule of wave crest, the image processing and monitoring algorithm of gap fault and overlapping fault are obtained by analyzing the position information of wave peak. A linear fitting algorithm is designed to predict the position of the peak of optical fiber wave. The experimental results show that the prediction algorithm is effective. Finally, the automatic dialing control of fiber winding machine is finished. A multi-axis motion controller based on ARM and FPGA is designed to replace the original PLC control scheme. The flexibility of the control system is improved. Experiments show that the multi-axis synchronization controller has good synchronization control effect.
【学位授予单位】：北方工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN96;TN253;TP391.41

【参考文献】