Girolami, M; Findlay, J; (1998) An unsupervised artificial neural network approach to adaptive noise cancellation applied to on-line tool condition monitoring. In: Adolfsson, J and Karlsen, J, (eds.) MECHATRONICS '98. (pp. 769 - 773). PERGAMON PRESS LTD
Full text not available from this repository.
Enhancing detected machining signals such as cutting force is an important research field in automatic manufacturing systems. For a milling machine, the problem is particularly difficult because of multiple sources caused by the numerous cutting teeth. The problem is further compounded in that the multiple sources are distorted and combined at the sensors in a non-stationary and unknown manner. Multivariable adaptive noise cancellation (MVANC) techniques have been employed to reduce the desired signal distortion; however, certain restrictions such as noise alone periods and their detection is necessary. This paper reports on the practical application of an unsupervised artificial neural network (ANN) model to this particular noise reduction and signal enhancement problem. This method allows the noise reduction to proceed in a 'blind' and unsupervised manner.
|Title:||An unsupervised artificial neural network approach to adaptive noise cancellation applied to on-line tool condition monitoring|
|Event:||6th UK Mechatronics Forum International Conference|
|Dates:||1998-09-09 - 1998-09-11|
|UCL classification:||UCL > School of BEAMS > Faculty of Maths and Physical Sciences > Statistical Science|
Archive Staff Only: edit this record