赵久强,王震洲

• 1:河北科技大学信息科学与工程学院

摘要(Abstract)：

针对温度会影响红外CO_2传感器的输出电压,造成对CO_2的浓度检测误差较大的问题,提出了一种基于L-M贝叶斯正则化BP神经网络的温度补偿方法。实验中将传感器输出电压比和温度作为神经网络的输入,CO_2浓度作为神经网络的输出,并通过L-M算法和贝叶斯正则化对神经网络进行优化。经过实验仿真证明,在温度补偿后红外CO_2传感器测量输出的浓度值最大相对误差为4.557 8%,具有较高的精确度。因此L-M贝叶斯正则化BP神经网络能对红外CO_2传感器进行有效的温度补偿,可为相关红外传感器仪器的改进提供参考。

关键词(KeyWords)： 计算机神经网络;红外CO2传感器;BP神经网络;L-M算法;贝叶斯正则化;温度补偿

Abstract:

Keywords:

基金项目(Foundation): 河北省科技支撑计划项目(16273705D)

作者(Author): 赵久强,王震洲

参考文献(References)：

• [1]张广军,武晓利.新型高性能红外二氧化碳传感器[J].红外与激光工程,2002,31(6):540-544.ZHANG Guangjun,WU Xiaoli.Novel IR carbon dioxide sensor with high performance[J].Infrared and Laser Engineering,2002,31(6):540-544.
• [2]涂志华,赵阳,郑力文,等.基于非分散红外(NDIR)技术的土壤剖面二氧化碳浓度的测定[J].光谱学与光谱分析,2015,35(4):997-1000.TU Zhihua,ZHAO Yang,ZHENG Liwen,et al.Detection of carbon dioxide concentration in soil profile based on non-dispersive infrared spectroscopy technique[J].Spectroscopy and Spectral Analysis,2015,35(4):997-1000.
• [3]闫玲,姚毅,昌学年,等.红外吸收法测定烟气中CO2浓度的研究[J].仪器仪表与分析监测,2009,100(4):27-29.YAN Ling,YAO Yi,CHANG Xuenian,et al.Study on infrared absorption measurement for the concentration of carbon dioxide in flue gas[J].Instrumentation Analysis Monitoring,2009,100(4):27-29.
• [4]琚雪梅,张巍,毕东云,等.红外吸收型CO2气体传感器的设计[J].传感器技术,2005,24(8):62-64.JU Xuemei,ZHANG Wei,BI Dongyun,et al.Design of infrared absorption CO2 gas sensor[J].Journal of Transducer Technology,2005,24(8):62-64.
• [5]王莉,陈小平.基于红外传感器的CO2测量装置[J].传感器与微系统,2011,30(11):115-117.WANG Li,CHEN Xiaoping.CO2measuring device based on infrared sensor[J].Transducer and Microsystem Technologies,2011,30(11):115-117.
• [6]崔颖,宋绍楼,彭继慎,等.关于瓦斯传感器的非线性校正的研究[J].微计算机信息,2010,26(25):138-139.CUI Ying,SONG Shaolou,PENG Jishen,et al.The study for non-linear correction based on gas sensor[J].Microcomputer Information,2010,26(25):138-139.
• [7]王建梅,覃文忠.基于L-M算法的BP神经网络分类器[J].武汉大学学报(信息科学版),2005,30(10):928-931.WANG Jianmei,QIN Wenzhong.BP neural network classifier based on Levenberg-Marquardt algorithm[J].Geomatics and Information Science of Wuhan University,2005,30(10):928-931.
• [8]赵映,陈小平.BP神经网络在红外CO2体积分数测量中的应用[J].传感器与微系统,2015,34(3):151-153.ZHAO Ying,CHEN Xiaoping.Application of BP neural networks in IR CO2 volume fraction measuring[J].Transducer and Microsystem Technologies,2015,34(3):151-153.
• [9]张鹏,杨兴锐,严翔,等.基于BP神经网络的风洞传感器非线性误差修正方法[J].兵工自动化,2016,35(3):31-34.ZHANG Peng,YANG Xingrui,YANG Xiang,et al.Nonlinear error correction method of wind tunnel sensor based on BP neural network[J].Ordnance Industry Automation,2016,35(3):31-34.
• [10]杨春生,牛红涛,隋良红,等.基于贝叶斯正则化算法BP神经网络钒电池SOC预测[J].现代电子技术,2016,39(8):158-161.YANG Chunsheng,NIU Hongtao,SUI Lianghong,et al.Application of BP neural network improved by Bayesian regularization algorithm in VRB SOC prediction[J].Modern Electronics Technique,2016,39(8):158-161.
• [11]刘刚,刘学仁,嵇英华,等.基于改进型BP神经网络的瓦斯传感器的非线性校正[J].传感器与微系统,2007,26(1):15-17.LIU Gang,LIU Xueren,JI Yinghua,et al.Non-linear Correction of methane Sensor based on improved BP neural network[J].Transducer and Microsystem Technologies,2007,26(1):15-17.
• [12]ROBERT F,THOMAS T.An automotive Bi-Source spectroscopic carbon dioxide sensor with pressure compensation[J].Sensors and Actuators B,2007,127(1):82-88.
• [13]KING T L,WEI M G,BREDA K,et al.Non-linear carbon dioxide determination using infrared gas sensors and neural network s with Bayesian regularization[J].Sensors and Actuators B,2009,136(1):242-247.
• [14]汪献忠,王晓东,赫树开.基于红外吸收型智能CO2浓度分析仪[J].仪表技术与传感器,2010,325(2):28-30.WANG Xianzhong,WANG Xiaodong,HE Shukai.Design of intelligent CO2 concentration analyzer based on infrared absorption[J].Instrument Technique and Sensor,2010,325(2):28-30.
• [15]陈明.Matlab神经网络原理与实例精解[M].北京:清华大学出版社,2013.
• [16]LAMPINEN J,VEHTARI A.Bayesian approach for neural networks:Review and case studies[J].Neural Networks the Official Journal of the International Neural Network Society,2001,14(3):257-274.
• [17]MACKAY D J C.Probable networks and plausible prediction:A review of practical Bayesian methods for supervised neural networks[J].Network Computation in Neural Systems,1995,6(3):469-505.
• [18]吴方良,石仲堃,杨向晖,等.基于L-M贝叶斯正则化方法的BP神经网络在潜艇声纳部位自噪声预报中的应用[J].船舶力学,2007,11(1):136-142.WU Fangliang,SHI Zhongkun,YANG Xianghui,et al.Submarine sonar self-noise forecast base on BP neural network and L-M Bayesian regularization algorithm[J].Journal of Ship Mechanics,2007,11(1):136-142.
• [19]OKUT H,WU X L,ROSA G J,et al.Predicting expected progeny difference for marbling score in Angus cattle using artificial neural networks and Bayesian regression models[J].Genetics Selection Evolution,2013,45(1):34.
• [20]VITO S D,PIGA M,MARTINOTTO L,et al.CO,NO2,and NOx,urban pollution monitoring with on-field calibrated electronic nose by automatic bayesian regularization[J].Sensors&Actuators B Chemical,2009,143(1):182-191.