基于数据融合技术的封闭鸡舍环境中多气体的检测
|
摘要
封闭鸡舍环境中存在多种有毒气体,常用的半导体气体传感器选择性差、受气体交叉干扰的影响大,导致测量误差大,不能满足检测需求。针对这个问题,采用经典数据融合方式中的最小二乘法和现代数据融合方式中的BP神经网络的4种训练函数,分别对多气体传感器输出信号进行仿真训练分析,并利用均方误差和迭代次数来评价仿真的性能。仿真和实验结果表明,有弹回的BP算法训练的网络性能最优,可有效地降低测量误差,平均相对误差和最大相对误差均在1%以内,满足封闭鸡舍环境检测需求。
There are a variety of toxic gases in the closed henhouse environment,but the poor selectivity of commonly used semiconductor gas sensor and the great effect of the cross coupling cause too large measurement error to meet the demand of measurement. Aiming at these deficiencies,the least squares method in the classical data fusion method and the 4 training functions of the BP neural network in the modern data fusion method to simulate and analyze the output signal of themulti-gas sensors are used to improve the measurement accuracy. The training square error and network iteration are adopted to evaluate the performance of the model. Simulation results show that the network performance trained by the rebound BP algorithm is optimal to reduce the measurement error effectively. The average relative error and the maximum relative error are limited within 1%to meet the requirements of environmental testing.
There are a variety of toxic gases in the closed henhouse environment,but the poor selectivity of commonly used semiconductor gas sensor and the great effect of the cross coupling cause too large measurement error to meet the demand of measurement. Aiming at these deficiencies,the least squares method in the classical data fusion method and the 4 training functions of the BP neural network in the modern data fusion method to simulate and analyze the output signal of themulti-gas sensors are used to improve the measurement accuracy. The training square error and network iteration are adopted to evaluate the performance of the model. Simulation results show that the network performance trained by the rebound BP algorithm is optimal to reduce the measurement error effectively. The average relative error and the maximum relative error are limited within 1%to meet the requirements of environmental testing.
引文
[1]RUDNITSKAYA A,LEGIN A.Sensor systems,electronic tongues and electronic noses,for the monitoring of biotechnological processes[J].Journal of Industrial Microbiology and Biotechnology,2008,35:443-451.
[2]朱文鹏.舍内小气候环境对肉仔鸡疾病发生的影响及改善措施研究[J].畜牧与饲料科学,2013,34(12):83-84.
[3]BRANCA A,SIMONIAN P,FERRANTE M,et al.Electronic nose based discrimination of a perfumery compound in a fragrance[J].Sensors and Actuators B,2003,92(1-2):222-227.
[4]WANG H F,MOON I,YANG S X,et al.A memetic particle awarm optimization algorithm for multi-modal optimization problems[J].Information Sciences,2012,197:38-52.
[5]JARADAT M A K,LANGARI R.A hybrid intelligent system for fault detection and sensor fusion[J].Applied Soft Computing,2009,9:415-422.
[6]邢广成,强天伟.人工神经网络的发展与应用[J].科技风,2012(15):65-66.
[7]殷士勇,王文中.基于FCM和BP神经网络的棉麻纤维识别方法研究[J].黑龙江大学自然科学学报,2013,30(3):405-409.
[8]徐佳伟.基于主成分分析法的BP神经网络在气敏传感器阵列检测方法中的应用[J].电力与能源,2015,36(2):154-157.
[9]周凤麒,吴微,熊焱.多输出单元BP神经网络梯度算法的收敛性[J].黑龙江大学自然科学学报,2006,23(1):108-113.
[10]梁喜凤,刘立豪.传感器阵列的CO与H2S混合气体检测系统设计与试验[J].中国计量学院学报,2015,26(1):39-45.
[11]杨守建,陈恳.BP神经网络性能与隐层结构的相关性探究[J].宁波大学学报(理工版),2013,26(1):48-52.
[12]PIZZILEO B,KANG L,GEORGE W,et al.Improved structure optimization for fuzzy-neural networks[J].IEEE Transactions on Fuzzy Systems,2012,6:1076-1089.
[13]龚雪飞,刘萍,简家文.基于PSO算法集成神经网络的多元有害气体检测系统[J].传感技术学报,2015,28(6):938-942.
[14]雷云涛.基于神经网络和冗余技术的传感器检测系统[J].电力系统及其自动化学报,2013,25(3):158-161.
[2]朱文鹏.舍内小气候环境对肉仔鸡疾病发生的影响及改善措施研究[J].畜牧与饲料科学,2013,34(12):83-84.
[3]BRANCA A,SIMONIAN P,FERRANTE M,et al.Electronic nose based discrimination of a perfumery compound in a fragrance[J].Sensors and Actuators B,2003,92(1-2):222-227.
[4]WANG H F,MOON I,YANG S X,et al.A memetic particle awarm optimization algorithm for multi-modal optimization problems[J].Information Sciences,2012,197:38-52.
[5]JARADAT M A K,LANGARI R.A hybrid intelligent system for fault detection and sensor fusion[J].Applied Soft Computing,2009,9:415-422.
[6]邢广成,强天伟.人工神经网络的发展与应用[J].科技风,2012(15):65-66.
[7]殷士勇,王文中.基于FCM和BP神经网络的棉麻纤维识别方法研究[J].黑龙江大学自然科学学报,2013,30(3):405-409.
[8]徐佳伟.基于主成分分析法的BP神经网络在气敏传感器阵列检测方法中的应用[J].电力与能源,2015,36(2):154-157.
[9]周凤麒,吴微,熊焱.多输出单元BP神经网络梯度算法的收敛性[J].黑龙江大学自然科学学报,2006,23(1):108-113.
[10]梁喜凤,刘立豪.传感器阵列的CO与H2S混合气体检测系统设计与试验[J].中国计量学院学报,2015,26(1):39-45.
[11]杨守建,陈恳.BP神经网络性能与隐层结构的相关性探究[J].宁波大学学报(理工版),2013,26(1):48-52.
[12]PIZZILEO B,KANG L,GEORGE W,et al.Improved structure optimization for fuzzy-neural networks[J].IEEE Transactions on Fuzzy Systems,2012,6:1076-1089.
[13]龚雪飞,刘萍,简家文.基于PSO算法集成神经网络的多元有害气体检测系统[J].传感技术学报,2015,28(6):938-942.
[14]雷云涛.基于神经网络和冗余技术的传感器检测系统[J].电力系统及其自动化学报,2013,25(3):158-161.