An AQI Level Forecasting Model Using Chi-square Test and BP Neural Network

Along with the development of the industrialization, the pollution becomes more and more serious. Many cities are often shrouded in a pollution haze, which threatens seriously the health of the people. Therefore, it is very necessary to establish a scientific and effective air quality forecast model. In this paper, an AQI level forecasting model using chi-square test and BP neural network is established. The model is based on the monitoring data of air pollution obtained from Shijiazhuang air quality monitoring stations. Firstly the model uses chi-square test method to determine the influence factors. Secondly it uses these influence factors data to train the neural network. Finally, the test of the forecasting model is evaluated. The results show that: The AQI level forecasting model has higher forecasting accuracy, it improves the effectiveness and practicability, and can provide more reliable decision evidence for environmental protection departments.