Forecasting hourly intermittent rainfall by deep belief networks with simple exponential smoothing

Abstract

Accurate rainfall forecasting is essential and useful in planning and managing water resource systems efficiently. The intermittent rainfall patterns increase the difficulty of accurately forecasting rainfall values. Recently, deep learning techniques have been popular and powerful in the forecasting area. Thus, this study employs deep belief networks with a simple exponential smoothing procedure (DBNSES) to forecast hourly intermittent rainfall values in Taiwan. Weather factors are used as independent variables to forecast rainfall volume. The simple exponential smoothing data preprocessing procedure is used to deal with the intermittent data patterns. Other three forecasting models, namely the least squares support vector regression (LSSVR), the generalized regression neural network (GRNN), and the back propagation neural network (BPNN), are employed to forecast rainfall with the same data sets. In addition, genetic algorithms are utilized to determine the parameters of four forecasting models. The empirical results indicate that the developed DBNSES models are superior to the other forecasting models in terms of forecasting accuracy. In addition, the DBNSES can obtain smaller values of RMSE than the previous studies. Therefore, the DBNSES model is a suitable and effective way of forecasting rainfall with intermittent data patterns.