Abstract
Slit valves play an important role in semiconductor manufacturing, enabling creation and maintaining of a vacuum environment required for wafer processing. Due to the high volume of production in the modern semiconductor industry, slit valves could experience severe degradation over their lifetime. If maintenance is not applied in due time, degraded valves may lead to defects in finished products due to pressure loss and particle generation. In this paper, we propose methods for signal processing and feature extraction for analysis of slit valve vibration signals. These methods are then used to demonstrate the ability to reliably, accurately and efficiently distinguish between vibration patterns of each individual valve via a multi-class classification procedure. Furthermore, instantaneous time–frequency entropy of valve vibrations enabled long term monitoring of a slit valve in production, in spite of variations in valve speed and operations.






Similar content being viewed by others
Notes
Definitions of those properties, as well as mathematical constraints on the kernels that are necessary to achieve them are summarized in the seminal book by Cohen (1995).
Observe all Mahalanobis distances between training vectors in class \(\nu _i\) and all Mahalanobis distances among training vectors in class \(\nu _j\). Then \(d(\nu _i,\nu _j,l)\) is the maximum of those distances. In a way, this is a measure of intra-class localization feature l provides for classes \(\nu _i\) and \(\nu _j\).
Observe Mahalanobis distances from any training vector in class \(\nu _i\) to any training vector in class \(\nu _j\). Then \(d(\nu _i,\nu _j,l)\) is the minimum of those distances. In a way, this is a measure of inter-class separation feature l provides for classes \(\nu _i\) and \(\nu _j\).
From each of the 4 segments, we got 16 time-domain based features and 19 time–frequency domain based features, for each of the three directions of vibrations, as well as the vibration RMS. In addition, the feature set also included the movement times of both closing and opening motions, yielding 1122 features.
One should note that our classification method transformed this 50-class classification into 1225 pairwise classification problems, which were solved using the kNN classification algorithm.
Those areas correspond to the portions of valve travel where increases in entropies were observed.
References
Analog Devices. (2015). ADXL327 data sheet. http://www.analog.com/static/imported-files/data_sheets/ADXL327.pdf. Accessed 15 August 2015
Bao, J., & Spanos, C. J. (2001). A simulation framework for lithography process monitoring and control using scatterometry. In AEC/APC Symposium XIII, Abstract available via http://impact.berkeley.edu/archive/secure/archives/seminars/abstracts/Junwei100101.pdf. Accessed 15 August 2015
Cholettte, M., Celen, M., Djurdjanovic, D., & Rasberry, J. (2013). Condition monitoring and operational decision making in semiconductor manufacturing. IEEE Transactions on Semiconductor Manufacturing, 26(4), 454–464.
Coates, M., & Fitzgerald, W. (1999). Regionally optimised time–frequency distributions using finite mixture models. Signal Processing, 77(3), 247–260.
Cohen, L. (1995). Time–frequency analysis: Theory and applications (1st ed.). Englewood Cliffs: Prentice Hall.
Djurdjanovic, D., Ni, J., & Lee, J. (2002). Time–frequency based sensor fusion in the assessment and monitoring of machine performance degradation. In Proceedings of the 2002 ASME international mechanical engineering congress and exposition (IMECE) (pp. 15–22).
Duda, R. O., Hart, P. E., & Stork, D. G. (2000). Pattern classification (2nd ed.). London: Wiley.
Dunn, J. C. (1973). A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters. Journal of Cybernetics, 3(3), 32–57.
Facco, P., Bezzo, F., Barolo, M., Mukherjee, R., & Romagnoli, J. A. (2009). Monitoring roughness and edge shape on semiconductors through multiresolution and multivariate image analysis. AIChE Journal, 55(5), 1147–1160.
Flett, J., & Bone, G. M. (2016). Fault detection and diagnosis of diesel engine valve trains. Mechanical Systems and Signal Processing, 72, 316–327.
Fulton, S., & Kim, M. (2007). ISMI consensus on preventive and predictive maintenance vision Ver. 1.1. In: International SEMATECH manufacturing initiative (ISMI). http://www.sematech.org/docubase/document/4819ceng.pdf. Accessed 15 August 2015
Guan, T., Kuang, Y. C., Ooi, M., Cheah, X. G., Tan, Y. S., & Demidenko, S. (2011). Data-driven condition-based maintenance of test handlers in semiconductor manufacturing. In Proceedings of the 6th IEEE international symposium on electronic design, test and application (DELTA) (pp. 189–194).
Heng, R., & Nor, M. (1998). Statistical analysis of sound and vibration signals for monitoring rolling element bearing condition. Applied Acoustics, 53(1), 211–226.
Hong, S. J., Lim, W. Y., Cheong, T., & May, G. S. (2012). Fault detection and classification in plasma etch equipment for semiconductor manufacturing. IEEE Transactions on Semiconductor Manufacturing, 25(1), 83–93.
Hopfe, V., Sheel, D., Spee, C., Tell, R., Martin, P., Beil, A., et al. (2003). In-situ monitoring for CVD processes. Elsevier Journal on Thin Solid Films, 442(1), 60–65.
Kim, B., & May, G. S. (1997). Real-time diagnosis of semiconductor manufacturing equipment using a hybrid neural network expert system. IEEE Transactions on Components, Packaging, and Manufacturing Technology—Part C, 20(1), 39–47.
Jeong, J., & Williams, W. J. (1992). Kernel design for reduced interference distributions. IEEE Transactions on Signal Processing, 40(2), 402–412.
Jones, D. L., & Baraniuk, R. G. (1995). An adaptive optimal-kernel time-frequency representation. IEEE Transactions on Signal Processing, 43(10), 2361–2371.
Jong, W. R., & Lin, T.-W. (2007). Statistical process control for e-diagnostic prediction of cluster-tool equipment. In Proceedings of the 33rd annual conference of the IEEE industrial electronics society—IECON 2007 (pp. 2916–2921).
Kittler, J. (1975). Mathematical methods of feature selection in pattern recognition. International Journal of Man-Machine Studies, 7(5), 609–637.
Kreßel, U. H.-G. (1999). Pairwise classification and support vector machines. In B. Schölkopf & C. J. C. Burges (Eds.), Advances in kernel methods (pp. 255–268). New York: MIT Press.
Lee, D. E., Hwang, I., Valente, C. M., Oliveira, J., & Dornfeld, D. A. (2006). Precision manufacturing process monitoring with acoustic emission. International Journal of Machine Tools and Manufacture, 46(2), 176–188.
Lee, S. K., Kim, T. R., Lee, S. G., & Park, S. K. (2010). Degradation mechanism of check valves in nuclear power plants. Annals of Nuclear Energy, 37(4), 621–627.
Lin, Y. H., Lee, W. S., & Wu, C. Y. (2013). A novel signal processing approach for valve health condition classification of a reciprocating compressor with seeded faults considering time-frequency partitions. Journal of Marine Science and Technology, 21(5), 578–585.
Mahamad, A. K., & Hiyama, T. (2011). Fault classification based artificial intelligent methods of induction motor bearing. International Journal Innovative Computing, Information and Control, 7(9), 5477–5494.
Musselman, M., & Djurdjanovic, D. (2012). Time–frequency distributions in the classification of epilepsy from EEG signals. Expert Systems with Applications, 39(13), 11413–11422.
National Instruments Corporation. (2015). sbRIO-9636 OEM operating instructions and specifications. http://www.ni.com/pdf/manuals/373378d.pdf. Accessed 15 August 2015
Papandreou-Suppappola, A. (Ed.). (2002). Applications in time–frequency signal processing (1st ed.). Boca Raton: CRC Press.
Pichler, K., Lughofer, E., Pichler, M., Buchegger, T., Klement, E. P., & Huschenbett, M. (2016). Fault detection in reciprocating compressor valves under varying load conditions. Mechanical Systems and Signal Processing, 70, 104–119.
Raoux, S., Liu, K., Guo, X., & Silvetti, D. (1998). In-situ RF diagnostic for PECVD process control. In Proceedings of the materials research society (MRS), symposium II (Vol. 502, pp. 53–58). Cambridge University Press.
Saxena, A., & Saad, A. (2007). Evolving an artificial neural network classifier for condition monitoring of rotating mechanical systems. Applied Soft Computing, 7(1), 441–454.
Shadmehr, R., Angell, D., Chou, P. B., Oehrlein, G. S., & Jaffe, R. S. (1992). Principal component analysis of optical emission spectroscopy and mass spectrometry: Application to reactive ion etch process parameter estimation using neural networks. Journal of the Electrochemical Society, 139(3), 907–914.
Shen, C. W., Cheng, M. J., & Chen, C. W. (2011). A fuzzy AHP-based fault diagnosis for semiconductor lithography process. International Journal of Innovative Computing, Information and Control, 7(2), 805–815.
Tang, J., Dornfeld, D., Pangrle, S. K., & Dangca, A. (1998). In-process detection of microscratching during CMP using acoustic emission sensing technology. Journal of Electronic Materials, 27(10), 1099–1103.
Wang, Q. H., Zhang, Y. Y., Cai, L., & Zhu, Y. S. (2009). Fault diagnosis for diesel valve trains based on non-negative matrix factorization and neural network ensemble. Mechanical Systems and Signal Processing, 23(5), 1683–1695.
Wu, H., Chang, C., Chen, B., Lee, C., Chang, C., Ko, J., Zhou, M., & Liang, M. (2003). Fault detection and classification of plasma CVD tool. In Proceedings of the 2003 IEEE international symposium on semiconductor manufacturing (pp. 123–125).
Wuxing, L., Tse, P. W., Guicai, Z., & Tielin, S. (2004). Classification of gear faults using cumulants and the radial basis function network. Mechanical Systems and Signal Processing, 18(2), 381–389.
Yang, B. S., Hwang, W. W., Ko, M. H., & Lee, S. J. (2005). Cavitation detection of butterfly valve using support vector machines. Journal of Sound and Vibration, 287(1), 25–43.
Acknowledgements
This research is supported in part by the National Science Foundation (NSF) grant IIP 1266279. The content of this paper is solely the responsibility of the authors and does not represent the official views of the NSF.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Musselman, M., Xie, H. & Djurdjanovic, D. Nonstationary signal analysis and support vector machine based classification for vibration based characterization and monitoring of slit valves in semiconductor manufacturing. J Intell Manuf 30, 1099–1110 (2019). https://doi.org/10.1007/s10845-017-1308-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10845-017-1308-4