The Study On Optimal Dispatch Of The Town Water Supply System Based On PSO

With the rapid development of the domestic economy and people's living standards, the speed and scale of urban construction also are getting faster and faster, so that the urban water consumption is increased continually. In order to ensure the required supply of water demand and water pressure, a considerable amount of funds is spent on construction, reform and expansion of the water distribution system in the cities every year. At the same time, urban water supply management system is running with low levels, so that the water supply energy consumption has been increased significantly, and the network operating cost also is increased. Therefore, in order to reduce investment, save energy and ensure that water supply is safe, unified planning, rational layout and reasonable scheduling and management for water supply system should be researched, which is significant for the water supply system to improve the economic and social benefits. In this paper, a urban water supply network is took for example, for the problems and shortcomings on optimal dispatch of the existing water supply systems, the particle swarm optimization (PSO) and artificial neural network are adopted to do research on the forecast for water demand, water supply network analysis model and optimal dispatch model which are three main components of the water supply system. The main contents in the paper are as follows:1. The basic principle and algorithm process of particle swarm optimization were introduced. The optimization performance influences of the parameters were analyzed and how to make certain the fitness function and the parameters were discussed.2. The influence factors of the short-term water demand prediction were expatiated and the change regularity of the short-term water consumption was analyzed. General regression neural network (GRNN) model based on PSO was put forward, and by using the strong global search performance of PSO and the best approximation performance of GRNN, the prediction model was achieved reunification of efficiency and accuracy, which were carried out its verification and analysis by taking example for the water demand prediction on Xi'an.3. Inverse analysis model of frictional coefficients of pipelines based on PSO was put forward, and based on fact watched data accurate data parameters of pipelines would be acquired by using inverse analysis model which were important parameters for the water supply network microcosmic model to establish. Inverse analysis model were carried out its verification and analysis by taking example for a certain urban water supply network finally.4. The water supply network microcosmic model was founded.5. According to combination conditions of pumps in water pumping stations and the restrictions conditions of water head and flow, the two series optimal dispatch model were founded based on the water supply network microcosmic model by using PSO, and the two series optimal dispatch model were carried out its verification and analysis by taking example for a certain urban water supply network finally.6. In the last section of the paper, the whole research is summarized and the prospects to the further research are put forward.