Neural Network Sensor Fusion for Tool Condition Monitoring
暂无分享,去创建一个
The design and implementation of a neural network-based system combining the outputs of several sensors (acoustic emission, force and spindle motor current) for monitoring progressive tool wear in a single point turning operation is described. Multichannel autoregressive series model parameters and power spectrum amplitudes are used as inputs to the network. The objective of the system is to extend the range of machining conditions over which the system performs successfully. A basic architecture for multiple sensor systems is outlined. Results of recent research to implement a real-time monitoring system are presented.
[1] David Dornfeld,et al. Tool Wear Detection Using Time Series Analysis of Acoustic Emission , 1989 .
[2] Michael S. Black,et al. Integration of multiple sensors to provide flexible control strategies , 1986, Proceedings. 1986 IEEE International Conference on Robotics and Automation.
[3] J. Tlusty,et al. A Critical Review of Sensors for Unmanned Machining , 1983 .
[4] A. Wayne Whitney,et al. A Direct Method of Nonparametric Measurement Selection , 1971, IEEE Transactions on Computers.