Study On The Detection Of Harmful And Toxic Gas Sensor Array And Data Fusion

The composition and content of poisonous and harmful gas detection hasbeen widely used in chemical industry, environmental protection, gas alarm,aerospace and other fields. Commonly used semiconductor gas sensor has theadvantages of fast response and high sensitivity, but due to the nature of cross-sensitivity to be applied to a variety of gas detection. But in practical applicationbackground, most of the components of the measured medium are more complex.Therefore, how to utilize the existing semiconductor sensor technology toachieve a variety of mixed gas detection becomes very necessary.This paper designs a sensor based on AIN semiconductor MEMSmicrostructure of four arrays, according to its output characteristic, backendtesting system is designed. Detection system to be measured by the sensor arrayto obtain the perception of the mixed gas, and its high dimensional response afterthe collection and processing transmission first machine, and then using themulti-sensor data fusion technology for identification and calculation of mixedgas being measured, calculated and measured the volume of gas components inthe component.The system hardware treats STM32F107micro controller as the core,including heating circuit, signal transmitting circuit, sensor range switch circuit,power circuit, acquisition circuit and communication circuit, and uses thekeilMDK4.0to complete the preparation of the hardware circuit of theunderlying drivers. Software parts apply LabVIEW and MATLAB mixedprogramming technology. LabVIEW mainly realizes the PC serial portcommunication, data display and storage and interactive functions such as man-machine interface. and complicated neural network computing is realized byMATLAB. Finally, this paper selects the CO, SO2, NO2and Cl2gas are analyzed in theactual test. The test results show that this system achieves the cosine designrequirements: the qualitative analysis of a variety of gases’ detectionsimultaneously and quantitative analysis of the content of each gas. Comparedwith traditional detection methods, this system realizes the on-line detection ofmixed gas quickly and accurately. It has certain practical significance on the fieldof gas leak detection and environment monitoring and so on.