Research on Gas Concentration Prediction Based on Wavelet Denoising and ARIMA Model

Abstract

In order to improve the reliability and accuracy of mine gas concentration prediction, a prediction model based on wavelet noise reduction and autoregressive differential moving average model (ARIMA) is proposed. the original data is decomposed, thresholded and reconstructed, and the noise in the time series data is stripped, and then the ARIMA module of Python is called to build a prediction model to fit the prediction data, The ARIMA (2,1,1) model parameters were selected to fit the best prediction model, and the prediction effect was tested. Research shows that the method based on wavelet noise reduction and ARIMA prediction model can effectively improve the prediction accuracy and reliability of gas concentration prediction in the short-term. The prediction results of this algorithm are compared with other prediction models. The prediction model can not only reflect the change trend of gas emission concentration, but also has high fitting effect and prediction accuracy.