An Improved Artificial Neural Network Method for Monitoring and Diagnosis of Engineering Processes With Applications

Abstract

An artificial neural network (ANN) is a common method used for the monitoring and diagnosis of engineering processes. The use of ANN consists of two phases: training and reasoning. ANN requires supervised training, which involves designing its structure (select the number of layers and number of nodes at each layer) and setting up its target outputs. In general, the structure design is done by trial and error, and the target outputs are set as "1-0" (1 means a specified process condition occurs and 0 means otherwise). This article demonstrates that the performance of ANN can be improved by adopting proper structure and by using the mean of the training samples as the target outputs. The proposed method is based on the idea that the best ANN is the one being most similar to the training samples or, equivalently, the one minimizing the entropy of the training samples. A procedure of using ANN for monitoring and diagnosing engineering processes is developed accordingly. Using three practical examples (tool condition monitoring in turning, machining condition monitoring in tapping, and metallographic condition monitoring in welding), it is shown that the proposed procedure improves the performance of ANN by at least 10%. In addition, the training time is also reduced.

Keywords

Artificial neural network (ANN)engineering monitoring and diagnosis ANN structure ANN target outputs tool condition monitoring in turning metallographic condition monitoring in welding

Get full access to this article

View all access options for this article.

References

Asoh, H. and Otsu, N. , 1989, "Nonlinear data analysis and multilayer perceptions ," Proceedings of the International Conference on Neural Networks (IJCNN) 1, 441-415.

Chauvian, Y. , 1989, "Principal component analysis by gradient descent on a constrained linear gebbain cell," Proceedings of the International Conference on Neural Networks (IJCNN) 1, 373-380.

Devijver, P.A. and Kittler, J. , 1982, Pattern Recognition: A Statistical Approach, Prentice Hall, New York.

Dornfeld, D.A. , 1991, "Neural networks sensor fusion for tool condition monitoring ," Annual of CIRP 39(1), 101-105.

Du, R. , Elbestawi, M.A. , and Li, S. , 1992, "Tool condition monitoring in turning using fuzzy set theory," Int. J. Mach. Tools Manufact. 32(6), 781-796.

Du, R. , Elbestawi, M.A. , and Wu, S.M. , 1995, "Automated monitoring of manufacturing processes: Part 1. Monitoring methods, Part 2: Applications," Trans. of ASME, J. of Eng. for Industry 117(2),133-141.

Fogel, D.B. , 1991, "An information criterion for optimal neural network selection ," IEEE Trans. on Neural Network 2(5), 490-497.

Gallinari, G. and Thiria, S. , 1988, "Multilayer perceptions and data analysis," Proceedings of the International Conference on Neural Networks (IJCNN) 1, 391-399.

Gutierrez, M. , Wang, J. , and Grondin, R. , 1989, "Estimating hidden unit number for two-layer perceptions ," Proceedings of the International Conference on Neural Networks (IJCNN) 1, 677-681.

10.

Kong, S.Y. and Hang, J.W. , 1988, "An algebraic projection analysis for optimal hidden units, size and learning rates in back-propagation learning," Proceedings of the International Conference on Neural Networks (IJCNN) 1, 363-370.

11.

Kramer, M.A. , 1991, "Nonlinear principal component analysis using autoassociative neural networks," AIChE 37,233-243.

12.

Li, C.J. and Yu, X.L. , 1993, "High pressure air compressor valve fault diagnosis using feedforward neural network," Proceedings of 1993 ASME Winter Annual Meeting 64, 375-380.

13.

Liu, T.I. and Lyons, C.S. , 1993, "Monitoring for glass production using neural networks ," Proceedings of 1993 ASME Winter Annual Meeting 64, 217-224.

14.

Rumelhart, D.E. and McClelland, J.L. , 1988, Parallel Distributed Processing: Explorations in the Microstructure of Cognition, 2 vols., MIT Press, Cambridge.

15.

Scharf, L.L. , 1991, Statistical Signal Processing, Detection, Estimation, and Time Series Analysis, Addison-Wesley, Reading, MA.

16.

Tansel, N. 1. and Mekdeci, C. , 1993, "Detection of tool failure in end milling with wavelet transformation and neural network," Proceedings of 1993 ASME WAM 64, 369-374.

17.

Unger, L.H. and Powell, B. , 1988, "Fault diagnosis using connectionest adaptation network ," in Proceedings of 1988 AIChE Annual Meeting , Washington, D.C.

18.

Venkatasubramanian and Chen , 1990, "Fault diagnosis using artificial neural network ," Proceedings of 1990 ASME Winter Annual Meeting 61, 110-128.

19.

Watanabe, K. , et al., 1989, "Incipient fault diagnosis of chemical processes via artificial neural network," AIChE 35(11), 1803-1812.