Design And Implementation Of Indoor Air Quality Inspection System Based On Gas Sensor Array

As an effective means of indoor gas environmental monitoring,air quality detection equipment has been widely used in recent years because of its portability,easy operation and real-time performance.At present,most of the traditional indoor air quality monitors just have a single function.Only one indoor environment value can be monitored,and the indoor air environment cannot be considered more comprehensively.On the other hand,indoor gas testing equipment does not analyze and process the data detected by gas sensor,neglecting the cross-sensitivity of the gas sensors.As a result,the data is influenced by the disturbing gas and it is difficult to achieve higher detection accuracy.Aiming at the above problems,a new type of indoor air quality detection system based on sensor array which has high precision and multi-function is designed and developed in this article.According to the specific detection target and gas sensor selection principles,in this paper,4 different gas sensors such as carbon dioxide sensor,oxygen sensor,carbon monoxide sensor and formaldehyde sensor is used to build a gas sensor array,the 4 sensors work together as a module to detect four indoor gas.This paper will select the 32 bit ARM Cortex-M4 core STM32F4 microcontroller as the main control chip,and TFTLCD module,Wi-Fi transmission module,gas sensor array module,PM2.5 detection module,temperature and humidity monitoring module,and other circuit,which together constitute the hardware platform of indoor air quality detection system.The main controller STM32F4 is responsible for the data reading,storage and forwarding of the main functional modules.The LCD module and Wi-Fi module provide data display and wireless transmission function for the detection system.The software of the detection system is composed of the collection system and the host computer.On the hardware of the system,the software program is designed to realize the collection of four kinds of indoor gas concentration data,the detection of PM2.5,the testing of temperature and humidity data,and real time display and wireless transmission.The computer client receives the indoor air quality which is wireless transmitted by the collection system to realize remote monitoring.In order to solve the low precision of the air quality detection system caused by the cross-interference of gas sensors,two algorithms are designed in this paper.One of the methods is to construct 4 BP neural networks with 3 layers.Through effective training,the relative error percentage error of the gas concentration data processed by the network is reduced by more than half compared with the original detection data.Another method is to establish a simplified gas cross-interference equation which based on the use of gas sensing mechanism of the gas sensor and the cross interference coefficients.The simplified cross-interference equation is obtained by function fitting,and it is used for data processing which can remove the cross-interference.As a result,the accuracy of the 4 gas concentration data is improved greatly.