A novel air quality index prediction model based on variational mode decomposition and SARIMA-GA-TCN

Abstract

Due to the rapid development of society and economy, air pollution often occurs. In order to reduce air pollution and improve the accurate prediction of air quality, this paper proposes a seasonal autoregressive comprehensive moving average model combined with the time convolutional network, and uses genetic algorithm and variational mode decomposition to find the parameters of the time convolutional network and reduce the randomness of the sequence. First, the seasonal autoregressive integral moving average model is used to predict the nonlinear information of the time series, and the difference between the original series and the first forecast is decomposed into variational modes, and then used as the prediction target of the time convolutional network. Then, the optimized time convolutional network is used for the second prediction to predict the linear information of the time series. To get the final prediction, add the predictions of the first model and the second model together. The experimental results show that the Root Mean Squared Error, Mean Absolute Percentage Error, Mean Absolute Error and Symmetric Mean Absolute Percentage Error of the model are smaller than those of other relevant prediction models, which proves that the model has the best prediction performance.