Prediction of Air Quality Combining Wavelet Transform, DCCA Correlation Analysis and LSTM Model

In the context of global climate change, air quality prediction work has a substantial impact on humans’ daily lives. The current extensive usage machine learning models for forecasting resulted in significant improvements to sector. long short-term memory network is deep model, which adds forgetting layer recurrent neural and several applications prediction. experimental data presented this research include pollution (SO2, NO2, PM10, PM2.5, O3, CO) meteorological (temperature, barometric pressure, humidity, wind speed). Initially, using calculate index (AQI) wavelet transform with adaptive Stein risk estimation threshold utilized enhance data. Using detrended cross-correlation analysis (DCCA), mutual association between elements then quantified. On short, medium, scales, model’s accuracy increases by 1%, 1.6%, 2%, 5% window sizes (h) 24, 48, 168, 5000, efficiency 5.72%, 8.64%, 8.29%, 3.42%, respectively. model developed paper improvement effect, its application forecast immense practical significance.