Authors: P. Sam Paul; D.S. Shylu; A.S. Varadarajan
Addresses: Department of Mechanical Engineering, Karunya Institute of Technology and Sciences, Coimbatore, 641114, Tamil Nadu, India ' Department of Electronics and Communication Engineering, Karunya Institute of Technology and Sciences, Coimbatore, 641114, Tamil Nadu, India ' M.E.S. College of Engineering, Malappuram, 679573, Kerala, India
Abstract: In metal cutting, tool wear is a critical parameter which can lead to machine down time, product discards and problems to human personnel. High cutting force, excessive cutting temperature and tool vibration of considerable magnitude form indications of tool wear. In this investigation, an attempt was made to fabricate a tool holder with inbuilt sensors that can sense these signals during turning of AISI 4340 steel having a hardness of 46 HRC using multicoated hard metal inserts with sculptured rake face. The signals received from tool with inbuilt sensors were synthesised by an artificial neural network model that can be trained to predict tool flank wear as a fusion product during turning of hardened steel. Cutting experiments were conducted to check and test the experimental, predicted results. From the results, it was found that the signals obtained from the tool with inbuilt sensors matched well with the signals of the dynamometer and also the predictions of the fusion model developed matched well with the experimental results. This scheme is simple in construction, cost effective and holds promise as a means for tool condition monitoring during automated turning operations.
Keywords: cutting force; cutting temperature; tool vibration; tool wear; hard turning; artificial neural network; ANN; linear regression; sensor fusion.
International Journal of Advanced Mechatronic Systems, 2017 Vol.7 No.6, pp.368 - 377
Received: 28 Nov 2017
Accepted: 05 Aug 2018
Published online: 17 Apr 2019 *