一种微型柔性机械臂内窥式加工工具的运动姿态和材料去除量研究

发布时间：2018-06-05 19:13

本文选题：内窥式工具 + 柔性机械臂　；参考：《电子科技大学》2012年硕士论文

【摘要】：内窥技术在医学上是一种先进而有效的无损或微创诊疗手段；在工业上也是一种应用极广的无损检测手段，常用于探查人造腔穴、管道、容器内壁以及各种组合体内部的结构和工况。倘若赋予内窥式工具一定的机械加工能力，那么就可以在不拆解不破坏被测物的前提下，直接对其内部缺陷（如破损、裂纹、毛刺等）进行修复处理，将“检”与“修”合为一步完成，并去掉“拆”和“装”这两个费时费力的无用功环节，将无损检测与修复彻底整合起来。材料去除量是机械加工过程中最重要的指标之一。以打磨修复喷气式引擎涡轮叶片上的一处细裂纹为例，为确保修复后的涡轮能在高速旋转时保持平衡，须精确获知修复过程中所去除材料的质量。然而在柔性内窥式工具的工作姿态和受力均未知的情况下，很难准确地计算出加工过程中的材料去除量。同时，柔性内窥式工具能否有效抑制材料去除过程中的有害振动也是决定它能否胜任修复任务的关键因素之一。本文正是针对上述问题，以一款含有微型柔性机械臂结构的内窥式加工工具（在PENTAX ES-3801医用内窥镜基础上改造而来）为主要研究对象，以假想的喷气式引擎涡轮叶片裂纹修复作为内窥式（研磨）加工的主要应用场景，在深入分析柔性机械臂的机械结构和运动学特性的基础上，对其在静力平衡条件下的工作姿态和加工作用力，以及加工时的动态稳定性和材料去除量的测量、估计和预测做了创新性和探索性的研究。主要内容为： 1.为获取柔性机械臂的工作姿态，研究了现有的运动姿态测量技术和系统的特点与不足，提出并搭建了基于机器视觉的可见光非侵入式运动姿态实时测量系统。针对可见光中的高光干扰，提出了一种改进的色键过滤补偿算法以提高该系统在非理想光照条件下的测量精度。使用二维相机校准样本对该测量系统的光路误差进行了校准并对其测量精度进行了评估。结果表明该系统在150mm见方的视场中的二维测量精度优于1mm；在同时跟踪15个运动目标时，系统的采样率可达25～30fps，达到了预期的测量需求。使用该系统测量了机械臂的运动姿态，并将测量结果作为后续研究的参照和验证标准。 2.由于受到微型柔性机械臂自身尺寸以及工作环境的限制，在内窥式加工过程中无法通过现有方法和技术对机械臂的工作姿态进行直接测量。在详细研究机械臂的结构和运动学特性对其工作姿态的影响后，建立了自由条件下和静力平衡条件下的机械臂运动姿态估计模型和作用力估计模型，并提出了一种基于模型的间接测量方法，，将确定机械臂运动姿态所需的十个参变量减化为两个，并通过一个LVDT位移传感器和一个微型载荷传感器测出。通过实验对模型的输出精度进行了评估。结果显示在常工作空间内，机械臂位置估计误差小于2mm，角度估计误差在±4°以内。在1～6N范围内，作用力估计误差小于0.5N。 3.研究了微型柔性机械臂的动力学特性。利用再生理论和稳定性叶瓣图对内窥式加工过程中可能出现的颤振现象进行了分析。提出了一个可用于在线式振动自动检测的相对能量指标R。利用该指标可以发现内窥式研磨过程中的早期振动。实验结果表明，R指标对加工过程中的振动现象有着非常高的辨识灵敏度，对于稳定工作时的加工参数变化却不敏感，具有较好的鲁棒性。 4.研究了柔性内窥式研磨加工过程中的能量分配和损耗，通过加工比能耗建立了材料去除量的分析经验估计模型。通过实验数据确定了该模型中的待定系数。并通过另两组独立加工实验验证了该模型的可重复性和输出精度。实验结果表明该模型输出的估计值与实际称量值的变化趋势吻合，平均相对误差约为22%。综上所述，本文提出了对具有柔性机械臂结构的内窥式工具在机械加工过程中的工作姿态、作用力以及材料去除量进行定量估计的方法，估计结果具有合理的精度。该结果证明内窥式工具可以被应用到机械加工作业中。在突破了运动自由度耦合以及加工动力传递等技术障碍后，（柔性）内窥式工具将成为一种非常有效的无损（检测）加工手段。
[Abstract]:endoscopic technology is an advanced and effective means of non - invasive or minimally invasive diagnosis and treatment in medicine ;
In the industry , it is also a kind of nondestructive testing method which is widely used in probing artificial cavity , pipeline , inner wall of container and internal structure and working condition of various combinations . If the endoscopic tool is endowed with certain mechanical processing capacity , it can repair the internal defects ( such as breakage , crack , burr , etc . ) directly without breaking the object under test .

The removal of material is one of the most important indexes in the machining process . In order to ensure that the repaired turbine can keep balance during high - speed rotation , it is difficult to accurately calculate the amount of material removed during the process . However , it is difficult to accurately calculate the amount of material removed during the process .

In view of the above problems , an endoscopic tool ( modified on PENTAX ES - 3801 medical endoscope ) containing micro flexible mechanical arm structure is used as the main research object . Based on the analysis of the mechanical structure and kinematic characteristics of flexible manipulator , the working attitude and processing force of flexible manipulator are analyzed in depth , and the dynamic stability and material removal amount are measured , estimated and predicted . The main contents are as follows :

1 . In order to obtain the working attitude of flexible manipulator , a real - time measurement system of visible light non - invasive motion attitude based on machine vision is put forward and a real - time measurement system of visible light non - invasive motion attitude based on machine vision is put forward .
When tracking 15 moving targets simultaneously , the sampling rate of the system can reach 25 - 30 fps , and the expected measurement demand is achieved . The motion posture of the robot arm is measured by using the system , and the measurement result is taken as the reference and verification standard of the subsequent study .

2 . The mechanical arm motion attitude estimation model and the force estimation model can not be directly measured by the existing methods and techniques because of the size of the micro flexible robot arm and the limitation of the working environment . After the influence of the structure and the kinematic characteristics of the mechanical arm on the working attitude of the robot arm is studied in detail , ten reference variables required to determine the motion posture of the robot arm are reduced to two , and the output accuracy of the model is evaluated through an experiment . The result shows that the error of the position estimation of the mechanical arm is less than 2mm and the angle estimation error is within 卤 4 掳 . The force estimation error is less than 0.5N in the range of 1 to 6 N .

3 . The dynamic characteristics of micro - flexible manipulator are studied . A relative energy index ( R ) which can be used in the auto - detection of on - line vibration is presented . The results show that the R - index has very high recognition sensitivity to the vibration phenomenon in machining process .

4 . The energy distribution and loss in the flexible endoscopic grinding process are studied . The empirical estimation model of material removal is established by processing specific energy consumption . The accuracy of the model is determined by the experimental data . The results show that the estimated value of the model is consistent with the change trend of the actual nominal value , and the average relative error is about 22 % .

In conclusion , this paper presents a method to estimate the working attitude , the force and the amount of material removal in the machining process of the endoscopic tool with flexible mechanical arm structure . The result shows that the endoscopic tool can be applied to the machining operation . After breaking through the technical obstacles such as the coupling of the motion degree of freedom and the transmission of the processing power , the ( flexible ) endoscopic tool will become a very effective nondestructive ( detection ) processing means .
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH878;TH16

【参考文献】