The Research Of Tool Condition Monitoring Based On Multi-scale Principal Component Analysis

Difficult to cut materials possess the characteristics of high hardness and low thermal conductivity.When these materials are machining,the cutting force and temperature are very high and the wear of the cutting tool is severe.The tool condition affects the surface quality and precision of the work-piece.So it is necessary to monitor the wear state of the cutting tool in real time.But traditional pattern recognition methods are costly.They are not suitable for monitoring the real process of machining obviously.In this paper,a new tool condition monitoring method based on multi-scale principal component analysis(MSPCA)was proposed in this paper.Its model can be built only by samples under normal conditions.In this method,the training sample set of normal operational condition is decomposed into different scales using wavelet multi resolution analysis.Then principal component analysis(PCA)model of each scale was constructed to select significance scales.The statistical indices and the corresponding control limits are constructed to monitor the tool wear based on PCA.To test the effectiveness of the proposed method,super alloy 4169 milling experiment was carried out.Force and vibration signals during the machining process were collected simultaneously to depict the characteristics of the tool wear variation.Based on the extracted features,the tool wear monitoring is realized by MSPCA.The analysis results show that the monitoring accuracy of this method can reach 100% and meet the need of tool condition monitoring.Meanwhile,an online tool condition monitoring system was built based on the virtual instrument and the method in this paper.In order to verify the reliability of system,a titanium alloy drilling experiment was carried out.The monitoring result is very good and satisfactoryThe research result in this paper show that tool condition monitoring method based MSPCA can overcome the weakness of the traditional pattern recognition methods.And it is more suitable for monitoring the process of machining these difficult-to-cut materials.According to the online monitoring result of drilling process,it can be concluded that MSPCA is a kind of effective means to realize the tool wear monitoring.