Stable cutting zone prediction in CNC turning using adaptive signal processing technique merged with artificial neural network and multi-objective genetic algorithm

In metal cutting, surface quality and material removal rate are the key parameters investigated by several researchers. It has been already established that, at high-speed machining, tool chatter deteriorates the work-piece surface and effects the material removal rate too. Numerous crucial investigations have been carried out regarding the enhancement of these parameters considering tool chatter as a major thread. In the recent advancement, signal processing techniques are being used for suppression of chatter. Moreover, it has been found these advance techniques helps in predicting the actual nature of chatter. However, the chatter signal recorded during machining usually contain contaminations merged with actual signal. Hence, it becomes a task for researchers to rectify the signal and predict a suitable cutting zone that is capable of obtaining good surface finish with acceptable material removal rate. In the present work, ensemble empirical mode decomposition technique has been used to rectify the signal and optimal cutting zone has been predicted using the artificial neural network and multi-objective genetic algorithm. Machining in the obtained optimal zone will upsurge the productivity, by decreasing tool chatter and increasing material removal rate simultaneously. To validate the proposed methodology, experiments have been performed within the obtained optimal zone. The results indicate the effectiveness of the proposed methodology.