Design And Realization Of Air Pollutant Monitoring And Early Warning System

In recent years,air pollution has become more and more serious,and ozone pollution is particularly prominent.Ozone pollution will harm people’s health and the ecological environment.As the first step in the treatment of air pollution,real-time monitoring and effective prediction of changes in air pollutants are of great significance to environmental protection and people’s daily lives.At present,the measuring instruments of domestic weather stations are huge,difficult to maintain,and cumbersome to display data.In response to the above problems,this thesis designs an air pollutant monitoring and early warning system,and builds a BP neural network ozone concentration prediction model based on the improved genetic particle swarm optimization(GA-PSO)algorithm.The air pollutant monitoring and early warning system is composed of hardware and software.Firstly,the data acquisition circuit is designed to collect the pollutant concentration data and related meteorological data,and the microprocessor circuit is designed to process the collected data,and then the data is transmitted through NB-IOT communication module.Then write the monitoring node equipment related software programs,access and configure the cloud platform,build a web server,and finally realize the display of data and information.Finally,through the system test and data analysis,the relative error of the system monitoring data is less than 5%,which verifies the accuracy of the air pollutant monitoring and early warning system.Aiming at the problem that traditional physical models are difficult to predict short-term near-ground ozone concentration,this thesis proposes an improved GA-PSO algorithm to optimize BP neural network parameters,and builds an ozone prediction model based on the improved GA-PSO-BP neural network.Using the historical data of a national weather station in a certain area as a sample to be substituted into the model for training,a better prediction effect was obtained and comparisons proved that the improved GA-PSO-BP ozone concentration prediction model is better than the traditional GA-PSO-BP ozone prediction model.