Multivariate Multisite Model MV.MS. Reg for Water Demand Forecasting

Abstract EN

A new multivariate multi-site MV.

MS.

Reg model is developed in this research depended on regression analysis mixed with Auto regressive multisite Matalas model (AMMM)and used for water demand forecasting .This developed model was applied to Kerkuk city as a case study for long term forecasting of water demand for different types such as domestic demand, industrial, commercial and public demand.

This was done by dividing the city into four sites and dividing the total water demand in each site into three types of demand(domestic, industrial with commercial and public demand) .Each type of water demand in each site was analyzed by multivariate regression base then the cross correlation between this type of demand for the four sites were included in the model using multi-site Matalas model.

Many explanatory variables were concluded to be most effective factors affecting different types of demands such as monthly temperature, monthly evaporation ,number of residential units, number of industrial and commercial units and number of public units which were forecasted successfully using Stochastic weather generation (SWG) method.