A new tool wear monitoring method based on multi-scale PCA
- 301 Downloads
A multi-scale principal component analysis (MSPCA) method is presented to realize online tool wear monitoring of milling process. In this method, the training sample set of normal operational condition is decomposed into different scales using wavelet multi resolution analysis. The statistical indices and the corresponding control limits are constructed to monitor the tool wear based on principal component analysis (PCA). By integration of PCA with wavelet transformation, the accuracy and robustness of tool wear monitoring model can be improved greatly. To test the effectiveness of the proposed method, a Ti–6Al–4V 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 root mean square and kurtosis features, the tool wear monitoring is realized by MSPCA and PCA respectively. The analysis and comparison results show that MSPCA can produce higher accuracy in comparison with PCA.
KeywordsTool wear monitoring Multi-scale principal component analysis Milling process Wavelet transformation
This project is supported by National Natural Science Foundation of China (51175371 and 51420105007), National Science and Technology Major Projects (2014ZX04012-014), and Tianjin Science and Technology Support Program (13ZCZDGX04000) and (14ZCZDGX00021).
- Lachouri, A., Baiche, K., Djeghader, R., et al. (2008). Analyze and fault diagnosis by multi-scale PCA. In ICTTA 2008. 3rd International conference on information and communication technologies: From theory to applications, 2008. IEEE (pp. 1–6).Google Scholar
- Misra, M., Yue, H. H., Qin, S. J., et al. (2002). Multivariate process monitoring and fault diagnosis by multi-scale PCA. Computers & Chemical Engineering, 26(9), 1281–1293.Google Scholar
- Moradi, H., Vossoughi, G., Movahhedy, M. R., et al. (2013). Forced vibration analysis of the milling process with structural nonlinearity, internal resonance, tool wear and process damping effects. International Journal of Non-linear Mechanics, 54, 22–34.Google Scholar
- Padilla, M., Perera, A., Montoliu, I., et al. (2010). Fault detection, identification, and reconstruction of faulty chemical gas sensors under drift conditions using Principal Component Analysis and Multiscale-PCA. In The 2010 international joint conference on neural networks (IJCNN). IEEE (pp. 1–7).Google Scholar
- Pearson, K. (1901). On lines and planes of closest fit to systems of points in space. Philosophical Magazine, 2, 559–572.Google Scholar