Monitoring of tool status using intelligent acoustic emissionsensing and decision based neural network

摘要

Successful automation of manufacturing processes is must forproduction of high quality, customized and economical products demandedby todays customers. This calls for uninterrupted machining withdesirable process parameters. This can be ensured by continuousmonitoring of machining status, which is strongly influenced by thecondition of the cutting tool. In this paper, a new scheme is proposedand evaluated for intelligent tool status monitoring. This paperdescribes the possibility of sensor integration in a machining processthrough a neural network. Experiments were conducted to study theinfluence of flank wear on AE and cutting forces. This collected data isthen used as training patterns for the neural network. The decisionbased neural network is used to integrate this information andconsequently for deciding on the condition of the tool. The results showthat acoustic emission-cutting force based multi-sensory-monitoringmethodology classifies tool status correctly