Determining meteorological parameters affecting urban water consumption using fuzzy clustering algorithm (Case study: Gorgan city)

Today, the estimation of water consumption amount is considered as one of the most important and practical issues; the complexity and effect of various factors and parameters on water consumption caused that analytical and mathematical methods not be practical enough in this regard. The conducted studies indicate that meteorological parameters are considered as the most important group in the short-term estimating of water consumption amount. In this study using fuzzy clustering algorithm the nonlinear relationship between climatic parameters as input variables and urban water consumption amount as output variable was investigated and it was compared with the results of linear correlation between mentioned variables. For this purpose the measured data of meteorological parameters of Hashem Abad synoptic station which were collected daily and the amount of water distributed in the city network during Farvardin to Tir of 1391, have been used. Finally, an equation was proposed for short-term estimation of water consumption amount based on climatic parameters using linear and nonlinear regression methods. The results showed that urban water consumption amount indicates a better correlation with temperature mean and considering maximum temperature can increase the accuracy in data estimation. By considering a grade 2 equation among input and output variables the results also showed that by having Nash-Satklif 0.71 and 43/35 Lit/s error’s root mean squares, the water consumption amount will be obtained more precisely and certain than other equations.