Abstract
Diamond cutting-tool wear has a direct impact on the processing accuracy of the machined surface in ultra-precision diamond cutting. It is difficult to monitor the tool’s condition because of the slight wear amount. This paper proposed a hybrid deep learning model for tool wear state prediction in ultra-precision diamond cutting. The cutting force was accurately estimated and the wear state of the diamond tool was predicted by using the hybrid deep learning model with the motion displacement, velocity, and other signals in the machining process. By carrying out machining experiments, this method can classify diamond-tool wear condition with an accuracy of more than 85%. Meanwhile, the effectiveness of the proposed method was verified by comparing it with a variety of machine learning models.
Funding Information
  • National Key R&D Program of China (2020YFB2007600, TC200H02J, 51975522, 52035013, U1709206, 51821093, 2018C01065)
  • Ministry of Industry and Information Technology’s Manufacturing High-quality Development Project (2020YFB2007600, TC200H02J, 51975522, 52035013, U1709206, 51821093, 2018C01065)
  • National Natural Science Foundation of China (2020YFB2007600, TC200H02J, 51975522, 52035013, U1709206, 51821093, 2018C01065)
  • Science Fund for Creative Research Groups of National Natural Science Foundation of China (2020YFB2007600, TC200H02J, 51975522, 52035013, U1709206, 51821093, 2018C01065)
  • Zhejiang Provincial Key R&D Program of China (2020YFB2007600, TC200H02J, 51975522, 52035013, U1709206, 51821093, 2018C01065)

This publication has 28 references indexed in Scilit: