Regional Water Demand Prediction and Analysis Based on Cobb-Douglas Model

Currently, regional water demand is mainly predicted by prediction models and according to actual water demand time series. However, regional water demand is affected by many factors, and the existing methods neglect dynamic mutual-restriction relation of various water demand influencing factors and influence of these factors on water demand and cannot calculate contribution rate of each factor to water demand. To address this problem, this paper, by adopting Cobb-Douglas production function, has established a regional water demand prediction model based on Cobb-Douglas model, by which the contribution rates of the regional water demand influencing factors can be calculated. It is indicated by example of Zhuhai in China that this proposed model possesses such advantages as simple modeling and high prediction accuracy by comparing with support vector machine and back-propagation neural networks models.