Abstract
Abstract
The cutting force and the vibration signal of a computer numerical control (CNC) turning machine centre are detected for on-line tool wear monitoring. The feature elements are first extracted from the detected signals. The feature indices are obtained from the feature elements through data preprocessing. Six data fusion methods are used for integrating the feature indices to obtain the fusion indices. The obtained fusion indices are used as the input data of a neural network for online tool wear monitoring. The feasibility of coupling a neural network algorithm with different data fusion methods is investigated, based on the monitored data. The research results show that using a data fusion neural network in tool wear monitoring is feasible.