基于关联信息的阵列气体传感器故障诊断研究

Array Chemical Gas Sensor Failure Diagnosis and Location Based on Information Association

人工嗅觉系统的实际工作环境一般较为恶劣,因此对气体传感器进行故障诊断,提高系统可靠性是必须的.由于气体传感器的交叉敏特性,使得传感器阵列的输出信息具有冗余性,本文提出了利用气体传感器阵列的冗余关联特性来进行气体传感器故障检测与诊断的新方法.在该方法中,人工神经网络被用于气体传感器故障关联信息的检测与定位,实际使用表明该方法完全能够实现阵列气体传感器故障的在线诊断与定位,并可适用于其它类似的系统.

It is necessary to enhance the reliability of artificial olfactory system in bad working environment. Crossed sensitivity of array chemical gas sensors, results in redundancy the output information. We hence used the characteristics of redundancy to solve the problems of gas sensor failure diagnosis, where the artificial neural network is be used. The practice uses show the novel method can meet the needs of the on-line diagnosis and location of array chemical sensors.