A Sensor Grouping Method for Industrial Sensor Health Management

Abstract:

As many sensor networks have been deployed in industry monitoring area, the focus on sensor data quality has also increased. Sensor networks provide us with process details which we can utilize to help making decisions on process monitoring.In order to make meaningful decisions, the quality of the data produced by sensors must be validated. As we evaluate the status of a specific sensor, we may also regard the status of the related sensors. If a sensor’s data show some abnormal, but the sensors related to it didn’t, we may have much more confidence to believe that the sensor is malfunction. In our early study, the sensors grouping strategy is manual. In this paper, we proposed a sensor grouping algorithm, which combines both PCA decouple method and the K-means cluster method. Finally, a test has been made with real data from an oilfield.

• Periodical: Applied Mechanics and Materials (Volume 621)
• Pages: 271-276
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.621.271
• Online since: August 2014
• Authors: Chang Jie Zhang*, Yu Liu
• Keywords: K-Mean, PCA, Sensor Data Quality, Sensor Networks, Sensors Grouping

References

[1] Hawbani, A., Wang, X., Xiong, Y., and Karmoshi: Wireless Sensor Network Routing Based on Sensors Grouping. Wireless Sensor Network, 6, 8.

DOI: 10.4236/wsn.2014.61002

[2] Lin, C. S., Chen, C. C., and Chen, A. C.: Partitioning sensors by node coverage grouping in wireless sensor networks. In Parallel and Distributed Processing with Applications (ISPA), 2010 International Symposium on (pp.306-312). IEEE.

DOI: 10.1109/ispa.2010.59

[3] Jackson, J. E.: A user's guide to principal components (Vol. 587). John Wiley & Sons.

[4] Lee, C., Choi, S. W., and Lee, I. B.: Sensor fault identification based on time-lagged PCA in dynamic processes. Chemometrics and Intelligent Laboratory Systems, 70(2), 165-178.

DOI: 10.1016/j.chemolab.2003.10.011

[5] Xie, Y., Chen, X., and Zhao, J.: Adaptive and Online Fault Detection Using RPCA Algorithm in Wireless Sensor Network Nodes. InIntelligent System Design and Engineering Application (ISDEA), 2012 Second International Conference on (pp.1371-1374.

DOI: 10.1109/isdea.2012.735

[6] J. L. Gao, Y. J. Xu and X. W. Li: Weighted-median Based Distributed Fault Detection for Wireless Sensor Networks, Journal of Software, Vol. 18, No. 5, 2007, pp.1208-1217.

DOI: 10.1360/jos181208

[7] Liu, Y., Yang, Y., Lv, X., and Wang, L.: A Self-Learning Sensor Fault Detection Framework for Industry Monitoring IoT. Mathematical Problems in Engineering, (2013).

DOI: 10.1155/2013/712028