Neural Network Classification of Audible Sound Signals for Process Monitoring during Machining

Flexible manufacturing systems require monitoring systems allowing to overlook all operations. Several sensing systems such as cutting force and torque, motor current and effective power, vibrations, acoustic emission or audible energy sound have been analyzed in recent years. Audible sound signals emitted during machining processes is one of the most practical techniques. The aim of this work is to characterize the audible sound signals from milling processes with different cutting parameters as a first approach to the study of audible sound based monitoring systems. The classification of audible sound signal features for process condition monitoring has been carried out using graphical analysis and neural network data processing.