Research On Indoor Air Quality Monitoring System Based On Edge Computing And Its Prediction Model

With the continuous improvement of residents' requirements on the quality of life,indoor air,as an important factor that contacts with residents in long-term and greatly affects their health,has attracted more and more attention,so a perfect way for real-time monitoring and prediction is urgently needed.In view of the weak awareness of indoor air quality monitoring,backward monitoring methods,and lack of complete and mature solutions,this paper develops an end-to-end indoor air quality monitoring system based on IoT architecture,supplemented by edge computing and ARIMA prediction model,for real-time and efficient indoor air monitoring.The main research contents of this paper are as follows:(1)Based on the IoT architeclture and technology,a three-layer architecture indoor air quality monitoring system is developed,which includes an air-aware terminal,transmission protocol and device network,and cloud monitoring platform,realizes the real-time acquisition,transmission and cloud storage methods of indoor air parameters.(2)By using edge gateway,the edge computing method of network edge side is studied,the functions of multi-device access and management,data analysis and cleaning,information decision and rule calculation are realized,which can relieve the network transmission pressure of air monitoring system,improve the ability of edge decision and the real-time performance of monitoring.(3)Through the steps of sample stabilization,order determination and model establishment,evaluation and result prediction,model validation and prediction error analysis,the ARIMA prediction model with indoor C02 concentration is established,which can realize the prediction of indoor C02 concentration and improve the early warning ability of air monitoring system.The system is tested and verified by taking a build indoor as the actual monitoring environment,and realized that the indoor air data collected and transmitted to the gateway for edge calculation,after data processing,upload the data to the cloud for storage and analysis,then the prediction model is established by using the data to predict the environmental change.The results show that the whole system can monitor the indoor air environment and forecast the change in real time,with high efficiency and intelligence,so the whole system has high practical value.