Air quality forecasting with hybrid LSTM and extended stationary wavelet transform

Air quality measurements and forecasting is one of the most popular research topics in field sustainable intelligent environmental design, urban area development pollution control, especially for Asia developing countries, such as China. Deep learning (DL) technologies time series data forecasting, recurrent neural network (RNN) long short term memory (LSTM) network, have attracted extensive attentions recent years been applied to AQI forecasting. However, two problems exist literature. First, volatility causes difficulties singular DL models produce reliable results. Second, a history air-quality required training stage, which usually unavailable. A novel model that integrates extended stationary wavelet transform (ESWT) nested short-term (NLSTM) PM2.5 air proposed this study. The results show method outperforms state-of-art methods recently published works terms different error metrics, absolute error, R2, MAE, RMSE, MAPE.