Daily Water Demand Forecast And Research And Application On Optimal Layout Of Monitoring Points

City water supply system is the lifeblood of a city. The main task of water supply system is safely to transport water accorded with national standards to users. It is also to content the users'demand of water flow, water pressure and water quality. In the premise of meeting the above requirements, how to make the water supply system to safely and efficiently operate with minimum cost? How to maximum decrease the cost of water supply? These problems have become the most important issue of water supply enterprises. Based on the above problems, municipal daily water demands forecast and optimal layout of the monitoring points (water pressure, water flow, water quality) had been highly valued by experts and scholars.Factors affecting urban water demands include two categories. One is Socio- economic factor, the other is weather and climate factor (such as rainfall, Teenage, temperature). The municipal daily water demands forecast model were established by using the single exponential smoothing, adaptive single exponential smoothing, quadratic curve exponential smoothing, gray model, BP neural network method and daily water demands data(table 3-1 daily water demands of A District). The short-term daily water demands were predicted by C++ program. Prediction accuracy of each models were analyzed. And the advantages and disadvantages of each model were compared. Considering the various factors, BP neural network method is the optimal model of the urban daily water demand forecast.According to the method of monitoring points layout of water distribution network at home and abroad, and analysis of water distribution network model of A District, the water pressure monitoring points of the method based on reflecting the change of water pressure in the nodes of networks and the method based on reflecting the size distribution of water pressure in the nodes of networks zones were optimized arrangement; the water quality monitoring points of the method based on the index of node flow coverage and the method based on the current age of nodes were optimized arrangement; the water flow monitoring points of the method based on the flow change sensitivity of networks nodes and the method based on the effective monitoring field were optimized arrangement.