Towards the improvement of tool condition monitoring systems in the manufacturing environment

Tool wear monitoring by machine learning techniques and singular spectrum analysis

“…O'Donnel et al [21] underlines also the high level of noise in vibration and acoustic signals as an additional difficulty for tool condition monitoring.…”

Robustness improvement of tool life estimation assisted by a virtual manufacturing cell

“…[11]. The nature of these influences and their effect on the applied sensor signals cannot be accounted in TCM strategies, as there is no means to measure or detect the changes of these factors in FMS.…”

“…The influences of factors such as variation in cutting variables, workpiece composition and tool wear combined with the complex nature of the cutting operation, make the monitoring of milling operation very complex. A lot of research work has been conducted theoretically and experimentally to investigate the effects of these factors on the TCM features and decision making process [1,[10][11][12], but the nature of these influences and their effect on the applied sensors signals has, to a large extent, remained unaccounted for by typical monitoring strategies, as there is no means to measure or detect changes of these factors in FMS [11]. The reliability of TCM system is considerably degraded by the facts that the influences of the cutting parameters and variation of workpiece hardness on TCM features are more significant than that of cutting tool wear.…”

Tool wear estimation using an analytic fuzzy classifier and support vector machines

“…However, continuous quantification of tool wear parameters, especially in the case of more complex machining operations such as milling, is not possible without a multiple sensors/features approach and a permanent adaptation of TCM model structure. This is particularly significant in the context of different process disturbances and other varying process parameters (O'Donnell et al 2001) due to which the correlation between the wear feature and the tool wear level can significantly oscillate. A certain feature may well correlate with the tool wear parameter for some wear level classes, while poorly for others.…”