A Multi-step Prediction Method of Urban Air Quality Index Based on Meteorological Factors Analysis

With the development of science and technology, Industry, transportation other industries used to discharge a large number pollutants into atmosphere, which results in air pollution. When pollution become serious, it will do great harm human health. High-precision Air Quality Index(AQI) prediction is as important weather prediction. People could arrange traveling their life according highly precise results, so better protect own Considering lot complex factors, we choose several potential meteorological factors historical data precisely forecast AQI. The principal component analysis (PCA) introduced our method conduct dimension reduction on nine order reduce noise complexity model calculation, improves accuracy AQI result. Then after PCA are input multi-step based LSSVM train refine it. Finally, set up experiment with Experimental show that proposed this paper has over classical ARIMA generalization than well.