Intelligent acoustic emission sensing in machining for tool condition monitoring

As the trend of machining toward automation proceeds, the real time identification of the states of the cutting process and cutting tool on the machine becomes increasingly important. The success of such an automated machining system implies production must be implemented with smart sensor systems to make it perform better. This dissertation describes the development of an intelligent acoustic emission based sensing system for tool condition monitoring in untended machining operations.;The acoustic emission generated from the sliding contact of metal-metal pairs using pin-on-disk tests is first investigated. The influence of process variables such as loading condition, sliding speed and distances, on the AE generated from sliding contact is studied. The sensitivity and effectiveness of the AE technique in detecting the changes of interaction characteristics in sliding contact is demonstrated. The possibility of in-process detection of these two phenomena by using the AE technique is discussed.;A method for recognizing tool wear states in a turning operation from the integrated information of cutting force and acoustic emission signals is presented. The approach, which employs gradient adaptive lattice analysis and pattern recognition techniques, is fast and yields accurate recognition of tool wear states in a wide range of cutting conditions. The results showed that this approach detected a fresh or worn tool state with a high percentage of correct classification (over 90%).;It was shown that the normalized residual variance (NRV), a dimensionless feature extracted from the lattice filter scheme, proved to be a valuable index for tool wear monitoring in single-tooth as well as multi-tooth milling under a wide range of cutting conditions. The feasibility of using acoustic emission for individual abnormal insert detection in addition to overall performance evaluation in the multi-tooth milling operation was also investigated. Furthermore, a frequency domain feature based on the periodicity characteristics of the acoustic emission signal generated from multi-tooth milling operations was also investigated and proved to be effective for tool wear monitoring and chipped tool detection.;Finally, the decision tree method and the group method of data handling (GMDH), due to their self-organizing capability for sensor integration, diagnostic reasoning and decision making, were adopted for the recognition and prediction of the tool wear state in a turning operation using acoustic emission and cutting force signals. Their performance for tool wear detection was discussed.