Study of Hydrostatic Spindle Cutting Force Prognostics and Health System using Support Vector Machines

• Main Author: 吳育齊
• Other Authors: 陳明飛
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/9zn3h2

碩士 === 國立彰化師範大學 === 機電工程學系 === 107 === The purpose of this paper is to design and develop a health diagnosis system of a hydrostatic spindle. Firstly, the design specifications of the presented hydrostatic spindle is based on the technical data of a given ball spindle. The main characteristics of the presented hydrostatic spindle include a radial system stiffness of 98 N/μm, an axial stiffness of 226.9 N/μm and a system flow of 4.36 L/min. Secondly, the mathematical model of axial and radial cutting force is established by using engineering algorithm. The pressure sensor data is obtained to compute the theoretical cutting force. By using the load cell the experiment cutting force of the spindle is obtain. The cutting force error between the mathematical model and the experiment test is less than 4%. Finally, the pressure sensor data and the spindle force are trained by the support vector machine (SVM), and then the accuracy is judged by the presented test data. In this paper, the accuracy rate of the experiment result is 99%. The hydrostatic spindle cutting force prognostics and health system using support vector machines is accomplished