实测资料分析是研究小浪底水库减淤作用的主要手段之一。研究一场洪水的冲淤特性 ,需在历史洪水中找出一场或几场水沙及河道边界与之相近的洪水来类比 ,许多复杂处理人工往往难以进行。本次研究引进人工神经网络理论 ,建立了具有较高精度的下游河道人工神经网络模型 ,并成功地应用于小浪底水库运用方式研究中。 The measured data analysis is one of the main methods to study the sedimentation reduction of Xiaolangdi Reservoir. To study the scouring and silting characteristics of a flood, it is necessary to find one or several water and sediment in the historical floods and analogous flood waters with similar river boundaries. Many complex treatments are often difficult to carry out. The research introduced the theory of artificial neural network and established the artificial neural network model of downstream river with high precision and successfully applied in the research of Xiaolangdi Reservoir.