Air Quality Index Prediction Using Error Back Propagation Algorithm and Improved Particle Swarm Optimization

Abstract

As the latest evaluation standards of air quality released by the State Environmental Protection Department, the Air Quality Index (AQI) is influenced by sulfur dioxide (SO2), nitrogen dioxide (NO2), particulate matter with particle size below 10 microns (PM10), particulate matter (PM2.5), carbon monoxide (CO) and ozone (O3) in the air. The variation of AQI shows nonlinearity and complexity. In order to improve prediction accuracy, this paper proposes an air quality prediction model based on Error Back Propagation (BP) algorithm. The model is optimized by Particle Swarm Optimization (PSO) algorithm using dynamic inertia weight and experience particles. The experimental results show the improved PSO-BP model significantly reduces iteration time, effectively improves the prediction accuracy, and provides a new method for the AQI prediction.