A new approach for water demand forecasting based on empirical mode decomposition

Forecasting of water demand is becoming an essential tool for the design, management and modernization of water supply and distribution systems. In this paper, the hybrid ensemble learning paradigm integrating ensemble empirical mode decomposition (EMD) and several single forecasting methods such as Autoregressive Integrated Moving Average (ARIMA) and artificial neural networks (ANN) models are proposed to improve the accuracy of water demand forecasting. In the hybrid model, EMD is used to decompose original data into a finite and often small number of sub-series. The each sub-series is modeled and forecasted by a several single models. Finally the forecast of water demand is obtained by aggregating all forecasting results of sub-series. To assess the effectiveness and predictability of proposed models, monthly water demand record data from Batu Pahat city in Johor of Peninsular Malaysia, have been used as a case study. The result shows that EMD-ANN model yield better forecasts than the single ARIMA, ANN and EMD-ARIMA models on several criteria.