Design Of Indoor Formaldehyde Concentration Detector Based On Data Fusion

The current indoor decoration materials contain a large amount of formaldehyde.Excessive formaldehyde can cause great harm to human health.Therefore,timely and accurate detection of indoor formaldehyde concentration is of great significance to ensure the health of residents.Traditional formaldehyde detection instruments are not conducive to the popularization of formaldehyde concentration detection due to their strong professionalism,bulkiness,and long detection time.Although the new formaldehyde detector is easy to operate and low in cost,it is easily affected by environmental factors and has low detection accuracy.In order to make the detection of formaldehyde smaller,faster and more convenient and improve its detection accuracy,this paper designs an indoor formaldehyde concentration detector based on data fusion.This paper uses the STM32F103 chip as the main control chip to collect the temperature,humidity,ethanol concentration and other environmental parameters in the indoor environment that have a greater impact on the detection of formaldehyde,and use the multiple linear regression algorithm to correct the error of the formaldehyde concentration.The system is mainly researched from four parts: hardware,software,data processing and data fusion.Based on the principles of portability and modularity,the hardware is divided into five parts: control unit,data acquisition,signal conditioning,data display and power supply circuit,and the selection and design of each part are carried out.The software part of the system is divided into four parts:main program,data acquisition,data fusion and data display.Aiming at the data processing of outliers,this paper uses an improved mean filtering method to filter the data by software.Finally,a linear regression model is established through a linear regression algorithm to perform data fusion,correct the detection value of formaldehyde concentration,and display it on the display screen in real time.After testing and verification,the accuracy of the predicted value of formaldehyde concentration processed by the data fusion model has been greatly improved than that of a single formaldehyde sensor,achieved the expected design effect.