Metacognitive learning approach for online tool condition monitoring
- 98 Downloads
As manufacturing processes become increasingly automated, so should tool condition monitoring (TCM) as it is impractical to have human workers monitor the state of the tools continuously. Tool condition is crucial to ensure the good quality of products—worn tools affect not only the surface quality but also the dimensional accuracy, which means higher reject rate of the products. Therefore, there is an urgent need to identify tool failures before it occurs on the fly. While various versions of intelligent tool condition monitoring have been proposed, most of them suffer from a cognitive nature of traditional machine learning algorithms. They focus on the how-to-learn process without paying attention to other two crucial issues—what-to-learn, and when-to-learn. The what-to-learn and the when-to-learn provide self-regulating mechanisms to select the training samples and to determine time instants to train a model. A novel TCM approach based on a psychologically plausible concept, namely the metacognitive scaffolding theory, is proposed and built upon a recently published algorithm—recurrent classifier (rClass). The learning process consists of three phases: what-to-learn, how-to-learn, when-to-learn and makes use of a generalized recurrent network structure as a cognitive component. Experimental studies with real-world manufacturing data streams were conducted where rClass demonstrated the highest accuracy while retaining the lowest complexity over its counterparts.
KeywordsPrognostic health management Online learning Evolving intelligent system Lifelong learning Nonstationary environments Concept drifts
This work is fully supported by the NTU start-up grant. The fourth author acknowledges the Austrian research funding association (FFG) within the scope of the ’IKT of the future’ programme (Contract # 849962), as well as the Austrian COMET-K2 programme of the Linz Center of Mechatronics (LCM).
- Angelov, P. P. (2010). Fuzzily connected multimodel systems evolving autonomously from data streams. IEEE Transactions on Systems, Man, Cybernetics Part-B: Cybernetics, 40(4), 898–910.Google Scholar
- Burke L. I. (1989) Automated identification of tool wear states in machining processes : An application of self-organising neural networks. PhD Thesis, Department of Industrial Engineering and Operations Research, UC at Berkeley, USAGoogle Scholar
- Cheng, W. Y., & Juang, C. F. (2014). A fuzzy model with online incremental SVM and margin-selective gradient descent learning for classification problems. IEEE Transactions on Fuzzy Systems, 22(2), 324–337.Google Scholar
- Dimla, D. E. Sr. (1998) Multivariate tool condition monitoring in a metal cutting operation using neural networks. Ph.D. Thesis, School of Engineering, The University of WolverhamptonGoogle Scholar
- Ding, S. X. (2008). Model-based fault diagnosis techniques. Berlin: Springer.Google Scholar
- Elanayar, S. V. T., Shin, Y. C., & Kumara, S. (1990). Machining condition monitoring for automation using neural networks. In ASME’s winter annual meeting, monitoring and control of manufacturing processes, PED (Vol. 44, pp. 85–100).Google Scholar
- Elbestawi, M. A., & Dumitrescu, M. (2006). Tool condition monitoring in machining—Neural networks. In W. Shen (Ed.), Information technology for balanced manufacturing systems. IFIP international federation for information processing (Vol. 20, pp. 5–16). Boston: Springer.Google Scholar
- Kasabov, N. K., & Song, Q. (2002). DENFIS: Dynamic evolving neural-fuzzy inference system and its application for time-series prediction. IEEE Transactions on Fuzzy Systems, 10(2), 144–154.Google Scholar
- Karmathi, S. V. (1994) On-line tool wear estimation in turning through sensor data fusion and neural networks. Ph.D Thesis, Department of Industrial and Manufacturing Engineering, The Pennsylvania State University, USA.Google Scholar
- Lister, P. M. (1993) On-line measurement of tool wear. Ph.D. Thesis, Manufacturing and Machine Tools Division, Department of Mechanical Engineering, UMIST.Google Scholar
- Liu T. I., & Ko E. J. (1990) On-line recognition of drill wear via artificial neural networks. In ASME’s winter annual meeting, monitoring and control for manufacturing processes, PED (Vol. 44, pp. 101–110).Google Scholar
- Noori-Khajavi, A., & Komanduri, R. (1995). Frequency and time domain analyses of sensor signals in drilling-II. Investigation on some problems associated with sensor integration. International Journal of Machine Tools Manufacture, 35(6), 795–815.Google Scholar
- Oentaryo, R.J et al. (2011) Bayesian ART-based fuzzy inference system: A new approach to prognosis of machining processes. In Proceedings of the IEEE international conference on prognostics and health management (PHM), Denver, Colorado (pp. 1–10)Google Scholar
- Oraby, S. E., & Hayhurst, D. R. (1991). Development of models for tool wear force relationships in metal cutting. International Journal of Machine Tools Manufacture, 33(2), 25–138.Google Scholar
- Pao, Y. H. (1989). Adaptive pattern recognition and neural networks. Reading, MA: Addison-Wesley.Google Scholar
- Pratama, M., et al. (2014c). A novel meta-cognitive-based scaffolding classifier to sequential non-stationary classification problems, In Proceedings of 2014 international conference on fuzzy systems (pp. 369–376).Google Scholar
- Pratama, M., Er, M.-J., Li, X., Oentaryo, R. J., Lughofer, E., & Arifin, I. (2013). Data-driven modeling based on dynamic parsimonious fuzzy neural network. Neurocomputing, 110, 18–28.Google Scholar
- Pratama, M., Lughofer, E., Lim, C. P., Rahayu, W., Dillon, T., & Budiyono, A. (2016). pClass+: A novel evolving semi-supervised classifier, online and in-press. International Journal of Fuzzy Systems. doi: 10.1007/s40815-016-0236-3.
- Rangwala, S., & Dornfeld, D. (1990). Sensor integration using neural networks for intelligent tool condition monitoring. Transactions of the American Society of Mechanical Engineers, Journal of Engineering for Industry, 112(3), 219–228.Google Scholar
- Scheffer, C. (2004). Practical machinery vibration analysis and predictive maintenance. Oxford: Elsevier.Google Scholar
- Subramanian, K., Savitha, R., & Suresh, R. (2013) Zero-error density maximization based learning algorithm for a neuro-fuzzy inference system II. In Proceedings of IEEE conference on fuzzy system (Fuzz-IEEE), Hyderabad, India (pp. 1–7)Google Scholar
- Wood, D. (2001). Scaffolding contingent tutoring and computer-based learning. International Journal of Artificial Intelligence in Education, 12(3), 280–292.Google Scholar
- Xu, Y., Wong, K. W., & Leung, C. S. (2006). Generalized recursive least square to the training of neural network. IEEE Transaction on Neural Networks, 17(1), 19.Google Scholar
- Xue, L., Liu, C.J., Lin, Y., & Zhang, W.J. (2015) On redundant interface: Concept and design principle. In Proceedings of IEEE/ASME international conference on advanced intelligent mechatronics. July 8–11, Busan, South Korea.Google Scholar
- Xiong, S., Azimi, J., & Fern, X. Z. (2014). Active learning of constraints for semi-supervised clustering. IEEE Transactions on Knowledge and Data Engineering, 26(1), 43–54.Google Scholar
- Yager, R. R., & Filev, D. P. (1994). Generation of fuzzy rules by mountain clustering. Journal of Intelligent and Fuzzy Rules, 2(3), 209–219.Google Scholar
- Zhang, W. J., & Van Luttervely, C. A. (2011). Toward a resilient manufacturing system. CIRP Annals-Manufacturing Technology, 60, 469–472.Google Scholar
- Zhao, Y., Jian, S., Gupta, M. M., Moody, W., Laverty, W. H., & Zhang, W. J. (2016). Developing a mapping from affective words to design parameters for affective design of apparel products. Textile Research Journal. doi: 10.1177/0040517516669072.