Model and Information Technology for Hourly Water Consumption Interval Forecasting

Abstract

Theoretical foundation of information technology for short-term water consumption forecasting has been considered. Hourly water consumption has been modelled as a cyclostationary conditional linear random process. Using the model an approach for interval forecasting has been proposed based on periodical autoregressive time series and nonparametric periodic quantile estimations. The corresponding information technology has been developed and analysed. The results of implementation have been presented and can be used for enabling optimal short-term water supply management.