Application of Ensemble Classifier Based on Deep Belief Network in Gas Identification

doi:10.3969/j.issn.1000-3428.2016.10.054

Abstract

Abstract: In order to reduce the influence of signal drift on gas identification,an ensemble classifier model based on Deep Belief Network(DBN) is proposed.A single DBN classifier is trained by different periods of datasets and the obtained classifier is used to classify the datasets.By minimizing the classification error,the optimal integrated weights of each single classifier are obtained.Particle Swarm Optimization(PSO) is used to find the optimal weights and integrate all the classifiers to get the final gas identification results.The performance of the proposed method is compared with that of the uniform weighted DBN and the optimal Support Vector Machine(SVM) by using a sensor array composed of 4 gas sensors.Experimental results show that,this method can maintain higher classification accuracy in a long time.To a certain extent,the effect of signal drift on the classification results is suppressed.

Key words: gas sensor array, drift compensation, Deep Belief Network(DBN), ensemble classifier, gas identification, network training

摘要： 为降低信号漂移对于气体识别的影响,提出一种基于深度信念网络(DBN)的集成分类器模型。利用不同时段的数据集训练单个DBN分类器,将得到的分类器对数据集进行分类,通过使分类误差最小得出每个单一分类器的最优集成权重,采用粒子群优化寻找最优权重并对所有分类器进行集成得到最终的气体识别结果。使用由4种气体传感器组成的传感器阵列对该方法和均匀加权DBN、最优支持向量机方法进行性能对比。实验结果表明,该方法能在较长时间里保持较高的分类准确率,在一定程度上抑制了信号漂移对分类结果的影响。

关键词: 气体传感器阵列, 漂移补偿, 深度信念网络, 集成分类器, 气体识别, 网络训练

CLC Number:

TP273

WANG Chunxiang,LI Lihong,ZHANG Di. Application of Ensemble Classifier Based on Deep Belief Network in Gas Identification[J]. Computer Engineering, doi: 10.3969/j.issn.1000-3428.2016.10.054.

王春香,李丽宏,张帝. 基于深度信念网络的集成分类器在气体识别中的应用[J]. 计算机工程, doi: 10.3969/j.issn.1000-3428.2016.10.054.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.3969/j.issn.1000-3428.2016.10.054

http://www.ecice06.com/EN/Y2016/V42/I10/318

References

参考文献［1］Gardner J W,Bartlett P N.A Brief History of Electronic Noses［J］.Sensors & Actuators B:Chemical,1994,38(1-3):210-211. ［2］王磊,刘锦淮.基于气体传感器阵列的动态检测系统［J］.传感器与微系统,2008,27(3):92-93. ［3］Gardner J W,Bartlett P N.Electronic Noses:Principles and Applications［M］.Oxford,UK:Oxford University Press,1999. ［4］刘航,唐祯安.基于动态分类器集成的MEMS气体传感器阵列的气体定性识别方法［J］.传感技术学报,2013,26(12):1649-1654. ［5］黄小燕,赵向阳,方智勇.电子鼻在气体检测中的应用研究［J］.传感器与微系统,2008,27(6):47-49. ［6］丁晖,刘君华,申忠如.基于分布式多子网神经网络的多组分混合气体识别［J］.仪器仪表学报,2001,22(6):592-594. ［7］李琦,杨艳菊.基于人工神经网络的苹果气体识别方法研究［J］.传感器与微系统,2007,26(9):61-63. ［8］Vergara A,Vembu S,Ayhan T,et al.Chemical Gas Sensor Drift Compensation Using Classifier Ensembles［J］.Sensors & Actuators B:Chemical,2012,166(10):320-329. ［9］汪丹,张亚非.基于支持向量机算法的气体识别研究［J］.传感器与微系统,2005,24(2):19-21. ［10］葛海峰,林继鹏,刘君华,等.基于支持向量机和小波分解的气体识别研究［J］.仪器仪表学报,2006,27(6):573-578. ［11］Osindero G E H S.A Fast Learning Algorithm for Deep Belief Nets［J］.Neural Computation,2006,18(7):1527-1554. ［12］余凯,贾磊,陈雨强,等.深度学习的昨天、今天和明天［J］.计算机研究与发展,2013,50(9):1799-1804. ［13］Hinton G E.A Practical Guide to Training Restricted Boltzmann Machines［M］//Montavon G,Orr G,Rob K.Berlin,Germany:Springer,2012:599-619. ［14］Llobet E,Brezmes J,Vilanova X,et al.Qualitative and Quantitative Analysis of Volatile Organic Compounds Using Transient and Steady-state Responses of a Thick-film Tin Oxide Gas Sensor Array［J］.Sensors & Actuators B:Chemical,1997,41(1-3):13-21. ［15］朱晔.面向安全椭圆曲线参数的PSO优化选择［J］.安庆师范学院学报:自然科学版,2014,20(4):21-24. ［16］刘昕,余隋怀,初建杰,等.改进PSO的飞机驾驶舱组合排序人机布局优化［J］.计算机工程与应用,2015,51(7):1-6. 编辑顾逸斐

[1]	LIU Feifei, WU Zhongdong, DING Longbin, ZHANG Kai. Intrusion Detection Algorithm for AMI Based on Improved Online Sequential Extreme Learning Machine [J]. Computer Engineering, 2020, 46(9): 136-142,148.
[2]	PAN Liangchen, WU Xinran, YUE Kun. User Preference Modeling Based on Deep Belief Network and Latent Variable Model [J]. Computer Engineering, 2020, 46(5): 54-62.
[3]	ZHU Jing, WU Zhongdong, DING Longbin, WANG Yang. DDoS Attack Detection Based on DBN in SDN Environment [J]. Computer Engineering, 2020, 46(4): 157-161,182.
[4]	LI Nana, HU Jianjian, GU Junhua, ZHANG Yajuan. Application of Optimized Deep Belief Network Model in Talent Evaluation [J]. Computer Engineering, 2020, 46(2): 80-87,102.
[5]	SHU Fei, CHEN Tao, WANG Bin, YANG Huiting, LI Mingxuan. An Anomaly Identification Method for Power Grid Industrial Control System Based on DBN-RF [J]. Computer Engineering, 2020, 46(11): 35-41.
[6]	QIAN Qing,TANG Guizhong,ZHANG Guangsming,DENG Xin,YIN Haipei. Short term prediction of itemized building energy consumption based on AR-DBN [J]. Computer Engineering, 2019, 45(6): 290-296.
[7]	WANG Yang, WU Zhongdong, HUO Zhongcai. Intrusion Detection Algorithm Based on DBN-KELM [J]. Computer Engineering, 2019, 45(10): 171-175,182.
[8]	WEI Sizheng,LIU Houquan,ZHAO Zhikai. Research on Intrusion Detection Based on DBN-ELM [J]. Computer Engineering, 2018, 44(9): 153-158.
[9]	MAO Yonghua,DAI Zhaosheng,GUI Xiaolin. Bisection Search Method of Radio Channel Based on Deep Belief Network [J]. Computer Engineering, 2018, 44(7): 86-90.
[10]	DOU Fangzheng,SUN Hanchang,SUN Xian,DIAO Wenhui,FU Kun. Remote Sensing Image Change Detection Method Based on DBN and Object Fusion [J]. Computer Engineering, 2018, 44(4): 294-298,304.
[11]	WANG Zhaokai,LI Yaxing,FENG Xupeng,LIU Lijun,HUANG Qingsong,LIU Xiaomei. Personalized Information Recommendation Based on Deep Belief Network [J]. Computer Engineering, 2016, 42(10): 201-206.
[12]	XU Qingyong,JIANG Shunliang,HUANG Wei,LI Jing,XU Shaoping,YE Famao. Image Classification Algorithm for Deep Belief Network Based on Multi-feature Fusion [J]. Computer Engineering, 2015, 41(11): 245-252.
[13]	LIU Zhi-Gang, DU Juan, HU Shao-Hua, LI Fen-Che. Time Series Prediction Method Based on Process Neural Networks [J]. Computer Engineering, 2012, 38(5): 199-201.
[14]	XIAO Gong, CAO Mao-Dun, LI Fen-Che, WANG Hai-Yang. Process Neural Network Training Based on Piecewise Linear Interpolation [J]. Computer Engineering, 2011, 37(20): 211-212.

Please choose a citation manager

Content to export