Design And Implementation Of Dust Detection System Based On Internet Of Things

Environmental monitoring is an important basis for environmental protection.It is used to monitor and analyze water,atmosphere,soil,noise and visual environment through scientific method.This monitoring and analysis is not only an important method for current environmental protection,but also a complete record of changes in the process of environmental governance.Then,making scientific and effective remediation plans based on the current results.We also further promote the establishment of national or local laws for environmental protection.For the current practical scenarios,we design a real-time online dust monitoring system for construction sites,road maintenance.The system uses the 485 sensor to collect data and upload data to the STM32 motherboard.The collected data is compiled into the national standard HJ212-2017 protocol through the program and then transmitted in two ways: After being processed by DTU,it is sent to the server,monitoring platform and supervision department separately;The other uses RS232 communication technology to enable real-time data display on LED and LCD screens.This article introduces the selection details of sensors,wireless network data transmission units,LED and LCD screens in the system,as well as the design ideas and implementation of the system hardware circuit;It also introduces how to design and implement various communication and data transmission software of the dust monitoring system,in which an online monitoring platform based on VUE framework is built to realize remote monitoring of data.The system can monitor temperature and humidity,air pressure,PM2.5/10,TSP,noise and other environmental parameters.It also have the functions of real-time display,reporting and historical data query.Through the real-time test that is comparing the real-time data released by the equipment and the national standard the data measured by two equipment are basically consistent;By placing the equipment in a fixed place for a month,the data collected from this stability test shows that the conditions whether the system can meet the actual monitoring needs and control the environment.In addition,The system also built a LSTM prediction model to predict and analyze AQI.The root-mean-square error of the prediction model is 8.45,and the accuracy rate is 95.75%.The prediction result is better than RNN and CNN.