The ST-GRNN Cooperative Training Model Based on Complex Network for Air Quality Prediction

In recent years, air pollution forecasting has become an important reference for governments when formulating environmental policies. However, accurate prediction of regional air quality has become a challenge due to the sparse spatial distribution of atmospheric monitoring stations. To address this problem, this paper proposes a neural network cooperative training and prediction model called “ST-GRNN”. The model incorporates complex network, Extreme Learning Machine (ELM), Long Short-Term Memory Network (LSTM), and Generalized Regression Neural Network (GRNN) to identify spatio-temporal features and accurately predict regional air quality. Comparative experiments using real datasets to predict PM2.5 concentrations show that the accuracy of the ST-GRNN model outperforms other methods.