Combustible Gas Discrimination by Pattern Recognition Analysis of Responses from Semiconductor Gas Sensor Array

Abstract:

The qualitative and quantitative identification of combustible gas mixture cannot be realized by a single sensor. Therefore, a semiconductor gas sensor array was built up. The experimental parameters including the dynamic and static information of the sensors were selected. The qualitative and quantitative identification of combustible gas mixture are achieved by the artificial neural network. The results show that this method for the qualitative identification of the combustible gas mixture is completely correct. The highest false rate of the quantitative analysis is 0.38% and the average false rate of the quantitative analysis is 0.079%. Achieve a good qualitative and quantitative identification.

• Periodical: Applied Mechanics and Materials (Volumes 303-306)
• Pages: 876-879
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.303-306.876
• Online since: February 2013
• Authors: Ping Sun, Zhong Hua Ou, Xing Feng
• Keywords: Artificial Neural Network (ANN), Combustible Gas, Gas Sensor Array, Pattern Recognition

References

[1] A. A. Tomchenko, G. P. Harmer, B. T. Marquis. Sensors and Actuators B. Vol. 93 (2003), pp.126-134.

[2] B. T. Marquis, J. F. Vetelino. Sensors and Actuators B. Vol. 77 (2001), pp.100-110.

[3] P. Sun, X. Feng, M. He. Smart Materials and Nanotechnology in Engineering. Vol. 20 (2012), pp.24-29.

[4] K. Arshak, E. Moore, G. M. Lyons. Sensor review. Vol. 24 (2004), pp.181-198.

[5] M. Yu, S. Wang. Gas Sensor Array. Sensor Technology. (1995).

[6] L. Xu. Chemistry Metrology. Science Press, Beijing. (2004).