基于RBF神经网络的非色散红外SF<sub>6</sub>气体传感器

薛宇; 常建华; 徐曦

doi:doi:10.3788/gzxb20164507.0723001

光子学报, 2016, 45 (7): 070723001, 网络出版: 2016-08-18

基于RBF神经网络的非色散红外SF₆气体传感器

Non-dispersive Infrared SF₆ Gas Sensor Based on RBF Neural Network

论文大纲

薛宇 ^1,*常建华 ^1,2徐曦 ¹

作者单位

¹ 南京信息工程大学江苏省气象探测与信息处理重点实验室, 南京 210044

² 南京信息工程大学江苏省大气环境与装备技术协同创新中心, 南京 210044

摘要

利用波段为2～20μm的电调制红外宽谱光源和中心波长为3.95 μm及10.55 μm的双通道热释电探测器, 采用单光源双波长光路结构设计了一种新型SF6气体传感器.运用径向基函数神经网络对传感器在检测过程中因环境温度变化所带来的测量误差进行补偿, 结果表明：SF6气体传感器在环境温度10～35℃、气体浓度0～0.200%范围内的检测准确度小于±1.5%FS, 相对标准偏差为1.56%, 可以有效消除在测量气体浓度时环境温度变化引起的非线性影响.与传统经验公式法和温度控制法相比, 该方法具有良好的测量准确度和稳定性, 且无需增加硬件温度补偿模块, 有利于传感器的小型化和低成本设计.

Abstract

The ro-vibrational spectra of the gas molecules is located in mid-infrared waveband, and then the information of the gas type and its concentration can be detected with a high precision based on the non-dispersive infrared technology. In this paper, a SF6 gas sensor was designed with the optical structure of the double light path of a single light, by utilizing a 2～20 μm electrically modulated thermal radiation source and a dual wavelength pyroelectric detector with the central wavelengths of 3.95 μm and 10.55 μm. The method of a radial basis function neural network algorithm was proposed to compensate the detection error caused by the variation of the ambient temperature. The experimental results show that the detection accuracy of this sensor is less than ±1.5%FS within the ambient temperature range of 10℃ to 35℃ and the gas concentrations from 0 to 0.200%. The relative standard deviation is 1.56%. It can effectively eliminate the nonlinear effects caused by the environmental temperature changing in measuring the gas concentration. Compared with the traditional compensation methods with the empirical formula or the temperature control scheme, our method has a better measuring accuracy and stability. Moreover, by using this method, the gas sensor doesn't need any temperature control module, which is beneficial to miniaturize the device size and reduce its cost.

参考文献

[1] GONDAL M A, BAIG M A, SHWEHDI M H. Laser sensor for detection of SF6 leaks in high power insulated switch gear systems［J］. IEEE Transactions on Dielectrics and Electrical Insulation, 2002, 9(3): 421-427.

[2] CALAZA C, SALLERAS M, SABATE N, et al. A MEMS-based thermal infrared emitter for an integrated NDIR spectrometer［J］. Microsystem Technologies, 2012, 18(7-8): 1147-1154.

[3] JANE H, RALPH P T. Optical gas sensing: a review［J］. Measurement Science & Technology, 2012, 24(1): 111-123.

[4] YASUDA T, YONEMURA S, TANI A. Comparison of the characteristics of small commercial NDIR CO2 sensor models and development of a portable CO2 measurement device［J］. Sensors, 2012, 12(3): 3641-3655.

[5] HODGKINSON J, SMITH R, HO W O, et al. Non-dispersive infra-red (NDIR) measurement of carbon dioxide at 4.2μm in a compact and optically efficient sensor［J］. Sensors & Actuators B Chemical, 2013, 186: 580-588.

[6] PETER W, FRANZ S, KARL M, et al. Near- and mid-infrared laser-optical sensors for gas analysis［J］. Optics & Lasers in Engineering, 2002, 37(2-3): 101-114.

[7] 于鑫, 高宗丽, 宋楠, 等. 袖珍式红外瓦斯检测仪的设计与实验［J］. 光子学报, 2014, 43(1): 58-63.

YU Xin, GAO Zong-li, SONG Nan, et al. Design and experiment of pocket infrared gas detector［J］. Acta Photonica Sinica, 2014, 43(1): 58-63.

[8] ZHANG Y, GAO W, SONG Z, et al. Design of a novel gas sensor structure based on mid-infrared absorption spectrum［J］. Sensors & Actuators B Chemical, 2010, 147(1): 5-9.

[9] 韩力群. 人工神经网络理论、设计及应用［M］. 化学工业出版社,2011.

[10] WANG Hai-rong, ZHANG Wei-you, LIU Dong, et al. Back propagation neural network model for temperature and humidity compensation of a non-dispersive infrared methane sensor［J］. Instrumentation Science & Technology, 2013, 41(6): 608-616.

[11] KUHN K, PIGNANELLI E, SCHUTZE A. Versatile gas detection system based on combined NDIR transmission and photoacoustic absorption measurements［J］. IEEE Sensors Journal, 2013, 13(3): 934-940.

[12] 乔俊飞, 韩红佳. RBF神经网络的结构动态优化设计［J］. 自动化学报, 2010, 36(6): 865-872.

QIAO Jun-fei, HAN Hong-jia. Optimal structure design for RBFNN structure［J］. Acta Automatica Sinica, 2010, 36(6): 865-872.

[13] 刘永平, 王霞, 李帅帅, 等. 基于红外技术的气体浓度检测方法研究［J］. 光子学报, 2015, 44(1): 011202.

LIU Yong-ping, WANG Xia, LI Shuai-shuai, et al. Gas concentration detection method based on infrared absorption spectroscopy technology［J］. Acta Photonica Sinica, 2015, 44(1): 011202.

[14] 荆耀秋, 江毅, 肖尚辉, 等. 一种差分吸收式光纤瓦斯传感系统［J］. 光子学报, 2014, 43(4): 0428002.

JING Yao-qiu, JIANG Yi, XIAO Shang-hui, et al. A differential absorption based optical fiber methane gas sensing system［J］. Acta Photonica Sinica, 2014, 43(4): 0428002.

[15] 张学典, 王业生, 伍雷, 等. 基于非色散红外CO2浓度测量的温度补偿研究［J］. 激光与红外, 2015, 45(4): 412-415.

ZHANG Xue-dian, WANG Ye-sheng, WU Lei, et al. Research on temperature compensation for CO2 concentration measurement based on NDIR［J］. Laser & Infrared, 2015, 45(4): 412-415.

[16] ZHANG W, TAN Q, LIU J, et al. Two-channel IR gas sensor with two detectors based on LiTaO3 single-crystal wafer［J］. Optics & Laser Technology, 2010, 42(8): 1223-1228.

[17] PARK J, YI S. Temperature compensated NDIR CH4 gas sensor with focused beam structure［J］. Procedia Engineering, 2010, 5(3): 1248-1251.

[18] TAN Q, ZHANG W, XUE C, et al. Design of mini-multi-gas monitoring system based on IR absorption［J］. Optics & Laser Technology, 2008, 40(5): 703-710.

[19] 曹培江, 彭双娇, 韩舜, 等.ZnO纳米/微米结构传感器对乙醇气敏性研究［J］.发光学报2014, 35(4): 460-464.

CAO Pei-Jiang, PENG Shuang-jiao, HAN Shun, et al. Gas-sensitive property of nano/micro-structured ZnO sensors on Ethanol［J］. Chinese Journal of Luminescence, 2014, 35(4): 460-464.

薛宇, 常建华, 徐曦. 基于RBF神经网络的非色散红外SF₆气体传感器[J]. 光子学报, 2016, 45(7): 070723001. XUE Yu, CHANG Jian-hua, XU Xi. Non-dispersive Infrared SF₆ Gas Sensor Based on RBF Neural Network[J]. ACTA PHOTONICA SINICA, 2016, 45(7): 070723001.

基于RBF神经网络的非色散红外SF₆气体传感器

关于本站 Cookie 的使用提示

全站搜索

基于RBF神经网络的非色散红外SF6气体传感器

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索

基于RBF神经网络的非色散红外SF₆气体传感器