Air Quality Prediction Based on Neural Network Model of Long Short-term Memory

Air pollution is a serious environmental problem, which has caused more and more concern all over the world. At present, research focuses on air quality prediction, usually using the following two methods: deterministic method and statistical method. Due to the unreliability of pollutant emission data and incomplete theoretical basis, the prediction accuracy of simulation results of these two methods is low, and these methods do not take into account the problem of time loss. In order to solve the problems of low prediction accuracy, low efficiency, and missing time factors in current air quality prediction research, a simple air quality prediction method LSTM neural network model is proposed. According to the characteristics of time series, the problem of multiple input time variables can be well solved. The datasets in two cities respectively prove that LSTM neural network model can effectively improve the accuracy of air quality prediction. The model can also be used to predict other multivariable input time series. Although air quality prediction based on LSTM presented in this paper has achieved the expected goal, there is still room for improvement in considering both spatial correlation and temporal dimension.