Automatic Measuring Process by Multi-level-sensing System using Optical Fibers and Artificial Intelligence

本计画的目的是利用光纤与人工智慧之多位准感测系统，发展自动量测程序。研究了多对光纤传感器的性能与测量精度和光纤位置的关系。考虑光纤的位置对测量系统的精度进行了仿真，并采用多级传感系统改变测量精度，以缩短测量时间。结果可以得出以下结论.从光纤传感器的基本特性分析，多对光纤的测量系统上级单对光纤的测量系统。通过对测量结果的比较发现，测量精度主要取决于光纤的定位误差和被测物体的反射系数.利用坐标测量机对多电平传感系统的性能进行了检验，该传感器具有足够的灵敏度，适用于粗测量过程.为了开发实际应用技术，在CNC控制的三坐标测量机上实现了自动工艺设计。从结果来看，需要讨论的是如何将测量数据中的距离和表面质量分离.通过对实验数据的计算，得到了模拟结果与实际工艺的充分吻合。将人工智能应用于路径测量的决策过程是非常重要的。

The aim of this project is to develop the automatic measuring process by multi-level-sensing system using optical fibers and artificial intelligence. The performance of the sensor with several couple of optical fibers were inves-tigated in relation to the measuring accuracy and the position of fiber. The accuracy of this measuring system was simulated considering the location of fibers and the multi-level-sensing system were carried out to alter the measur-ing accuracy in order to shorten the measuring time. As a result the following conclusions can be drawn.1. Analyzing a fundamental characteristics of optical fiber sensor, the measur-ing system with several couple of fibers is superior to that with single couple. Comparing the results, the accuracy maninly depends on the location error of attached fibers and the reflection factor of object to be measureed.2. Examining the performance of the multi-level sensing system using the coordi-nate measuring machine, the sensor propose here has enough sensitivity and is available to rough measuring process.3. To develop the practical application technique the automatic process planning was carried out with coordinate measuring machine which is controlled by CNC. From the results, it needs to discuss that how to separate the distance and sur-face quality in measured data.4. The sufficient agreement between the simulation and the practical process was obtained by calculating the several properties from the data. It is important to apply an aritificial intelligence to the decision process for measuring path.