Large-scale Irrigation On Water Resources Optimal Decision Model

This thesis studies the problem of water resource's optimal distribution in large irrigation areas basins. The problems of reducing dimension, of multipurpose optimal decision, and of integration of qualitative analysis and quantitative analysis for the large system decomposition-coordination step model are primarily calculated.1. This thesis puts forward a new model ?the model of crop's water rising and vaporizing regression nerve networks (NN). This model adds a deviation cell based on existing BP(NN) The effect of the deviation cell is to amend training errors of the NN, to increase convergence speed of the NN, and to reach expectant objective.2. Builds up decision model for water resource's optimal distribution of large irrigation areas in this thesis. This model resolves the optimal decision problem of how to distribute water resource to meet maximal total net profit in irrigation basins, when water supply potential of irrigation areas' canal head and bailing capability of pumping station hold the line. Two aspects of the problem are discussed: one is that lift irrigation is no need when abstraction volume capability of canal head can meet the demand; the other is that lift irrigation is in need when abstraction volume capability of canal head can not meet the demand.3. Puts forward matrix function form of the decomposition-coordination step model for water resource's optimal distribution in this thesis. It makes the model more prolific.4. The model of the thesis was examined by data of supplies and the simulation results fit well with practical data. So the model was useful to water resource's management of large irrigation areas.