Research On Detection Of Polluted Gas During Photocatalytic Degradation In Limited Space With Gas Sensor Array

With the improvement of living standards, the pollution with toxic and volatileorganic gas in the door is getting worse and worse, which directly threat the people’shealth. At present, the main methods of gas detection is conventional lab analysis, but theyare all expensive and huge volume which lead to the application difficulty or evenimpossible. While Electronic Nose owing to its simple, rapid, portable and low cost hasbeen widely applied to detect and analysis the gas/odor. In this paper, the stable metaloxide sensors were chosen to study the sensor array for the detection of the atmosphere.In order to study the appling the sensor array to detect the pollution air, a set ofphotocatalytic degradation and the gas detection device was designed. The device includedthree parts: the photocatalytic degradation unit, gas detection unit, and the gas circulationsystem. The whole process of photocatalytic degradation and the gas detection wascontrolled by LabVIEW program, which realized the totally automatic degradation anddetection of pollutant gas. Relevent experiments showed that the platform of this designedphotocatalytic degradation and sensor detection device had excellent airtightness andstability, and the experimental process and parameters had been set up.In the paper, this photocatalytic degradation and detection device was used to detectthe real-time process of P25/UV photocatalytic degradation of formaldehyde and toluene,and the data of Photoacoustic IR Multigas Monitor(1412) was treated as standard value ofgas concentration to access to analyze the sensor data. The results showed that the singlegas sensor through the learnning the characteristics and fitting method could get thepredict value with16%error, but the gas sensor array couldnot identify the concentrationof the air with this simple method. However, after the gas sensor array adopting BackPropagation Artificial Neural Network (BP-ANN) to learn corresponding relationship ofdata between sensor array and1412, this sensor array could individually predict thephotocatalytic degradation rate of any concentration of formaldehyde/toluene withP25/UV. The error of formaldehyde/toluene between prediction results and actual resultswere less than5%and10%, respectively. And this approach had very importantsignificance in practical application.