Water demand forecasting based on adaptive extreme learning machine

Abstract

Predicting water consumption is of key importance for water supply management, which is also relevant in processes for reviewing prices.In this study, a hybrid method based on extreme learning machine model with the adaptive metrics of inputs is proposed for improving forecasting accuracy. The adaptive metrics of inputs can solve the problems of amplitude changing and trend determination, and reduce the effect of the overfitting of networks. It was found that the proposed model is practical for water demand forecasting and outperforms the auto- regression (AR), artificial neural network (ANN), support vector machine(SVM) and extreme learning machine (ELM) models. Index Terms—Water consumption, Extreme learning machine, Forecasting, Time series