Wavelet Convolutional Neural Network for Forecasting Malaysian PM10 Time Series Data

Abstract

Hourly particulate matter time series data from eight air quality monitoring stations in Peninsular Malaysia were forecast by using the Convolutional Neural Network (CNN) algorithm. Instead of using the original time series, which are time-domain sequence data, this study used the time-frequency domain sequence data retrieved by wavelet transformation. Air pollutants' concentration considered for this study is the particulate matter with a diameter of 10 microns or less, PM 10 . The transformation used in this study is the Morlet wavelet transform, which is continuous wavelet transformation (CWT). Different time steps for the time series dependencies were considered to assess the PM 10 dependencies on its past values. The results were compared with the results from the CNN algorithm using the original time series. It is shown that the Wavelet Convolutional Neural Network algorithm improves the forecast accuracy of the PM 10 time series.