Integrated RAGA-PP water demand forecast model (case study: Shaanxi Province, China)

Abstract

The demand for water resources in Shaanxi Province increases greatly due to the continuous growth of its population and the rapid development of the social economy. Water demand forecasting is a significant issue in the designing, maintaining and operating of a reliable and economical water supply system. An explicit mathematical method was presented in this study, based on the indicators of industrial output value, such as the gross output value of agriculture, forestry, animal husbandry and fishery. The impact of total retail sales and year trends in the domestic or industrial water demands, can accurately forecast the water demand fluctuations for a municipality. Adopt RAGA-PP optimal selection model through a grouping method of data handling for water demand management to test for the case study Shaanxi, China. Results showed that the prediction effect of multivariate logarithmic model accuracy can reach 99.50%, and it is estimated that the demand for water resources in Shaanxi would exceed 10 billion m3 by 2030. The average relative error of the water consumption from 2015 to 2017 is 3.05% for the model of multiple linear and 0.50% for the model of multivariate logarithm model. Our framework can assist in developing sustainable solutions. HIGHLIGHTS Using the RAGA-PP model to realize the optimal selection of water resources demand indicators can improve model accuracy compared with model solving.; Adopt a curve fitting method and regression analysis method to establish the water demand model and solve it.; Due to the huge consumption of agricultural water, this paper puts forward corresponding agricultural water-saving measures.;

Keywords

• accelerated genetic algorithm
• projection pursuit model
• regression analysis
• water demand forecasting