Development of an intelligent multisensor chatter detection system in milling

The development of a chatter detection system based on multiple sensors and suitable for application in industrial conditions was investigated in this paper. The signals obtained from a monitoring system composed of accelerometers mounted on the machine head and an axial force sensor were processed by using advanced signal analysis techniques such as wavelet decomposition. The statistical parameters obtained from wavelet decomposition were used to detect chatter by using an artificial intelligence classification system based on neural networks. The outputs of the neural networks for each sensor signal were further combined by using different strategies in order to obtain a multisensor chatter indicator. The performances of different strategies were evaluated by using experimental data, evidencing that it is possible to obtain an efficient chatter detection system both in terms of accuracy and of robustness against malfunctions and compatible with modern machine tool operation and automation.