A procedure for spatial aggregation of synthetic water demand time series is presented. Starting from synthetic water demand time series generated at user level and reproducing mean and variance of the corresponding observed series, the procedure allows the aggregated series to preserve the statistics of interest observed at the aggregation level considered. The procedure uses a method proposed by Iman and Conover (1982): the synthetic user water demand time series are reordered to preserve the observed spatial correlations of appropriate lag-time; then, the spatial aggregation of these series leads to a series representative of the user group reproducing the corresponding means and variances for the different hours of the generic day. Four different ways of performing this aggregation have been investigated and compared. Application to a case study consisting of the water demands of 21 users highlights that the approaches considered show different levels of effectiveness in reproducing the statistics, but overall the procedure proposed is a valid tool for the bottom-up generation of synthetic water demand time series.
A procedure for spatial aggregation of synthetic water demand time series
ALVISI, Stefano;ANSALONI, Nicola;FRANCHINI, Marco
2014
Abstract
A procedure for spatial aggregation of synthetic water demand time series is presented. Starting from synthetic water demand time series generated at user level and reproducing mean and variance of the corresponding observed series, the procedure allows the aggregated series to preserve the statistics of interest observed at the aggregation level considered. The procedure uses a method proposed by Iman and Conover (1982): the synthetic user water demand time series are reordered to preserve the observed spatial correlations of appropriate lag-time; then, the spatial aggregation of these series leads to a series representative of the user group reproducing the corresponding means and variances for the different hours of the generic day. Four different ways of performing this aggregation have been investigated and compared. Application to a case study consisting of the water demands of 21 users highlights that the approaches considered show different levels of effectiveness in reproducing the statistics, but overall the procedure proposed is a valid tool for the bottom-up generation of synthetic water demand time series.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
