The development of an in-process surface roughness adaptive control system in turning operations |
| |
Authors: | Julie Z Zhang Joseph C Chen E Daniel Kirby |
| |
Affiliation: | (1) Department of Industrial Technology, University of Northern Iowa, Ceder Falls, IA, USA;(2) Industrial Technology Program, Department of Agricultural and Biosystem Engineering, Iowa State University, Ames, IA, USA |
| |
Abstract: | This research shows the development of an in-process surface roughness adaptive control (ISRAC) system in turning operations.
An artificial neural network (ANN) was employed to establish two subsystems: the neural network-based, in-process surface
roughness prediction (INNSRP) subsystem and the neural network-based, in-process adaptive parameter control (INNAPC) subsystem.
The two subsystems predicted surface roughness and adapted feed rate using data from not only cutting parameters (such as
feed rate, spindle speed, and depth of cut), but also vibration signals detected by an accelerometer sensor. The INNSRP subsystem
predicted surface roughness during the finish cutting process with an accuracy of 92.42%. The integration of the two subsystems
led to the neural-networks-based surface roughness adaptive control (INNSRAC) system. The 100% success rate for adaptive control
of the test runs proved that this proposed system could be implemented to adaptively control surface roughness during turning
operations. |
| |
Keywords: | Surface roughness Turning operations Adaptive control Neural-networks |
本文献已被 SpringerLink 等数据库收录! |
|