Based on the principle of linear and nonlinear Muskingum-Cunge scheme, a model is developed to solve the routing in long river with great lateral flow.

The paper approaches and analyzes the application of linear regression method,Muskingum method and method of confluence coefficient to the calculation of low flow of Ning-Meng section of the Yellow River based on the principle of hydrology of flow routing and taking a day as time step.

Then the general design flood with flood peak of 1132m3/s is gained by Muskingum method and sup.

Taking gradient and slope angle as main factors,a watershed is discretized into some sub-catchments according to natural draining divide,and aiming at characteristics of the research area,a rainfall-runoff distributed hydrological model orienting to catchment scale is established in this thesis with combination of runoff yielding and water routing model of SCS and Muskingum method.

With the application of Muskingum routing method,Muskingum water-level simulating method and diffusion wave nonlinear water-level method,flood is forecasted and regulated in the part basin between Wa.

In order to solve the problem of complexity and poor accuracy for parameter estimation in Muskingum model,the chaos was incorporated into particle swarm optimization,chaotic particle swarm optimization algorithm to estimate the parameter of Mustingum model was set up.