All kinds of random uncertainties on the impact of convergence calculation

Convergence system is a random system and the main sources of random uncertainty have model input (meteorological and hydrological data, etc.) model structure and model parameters and so on. Based on the model of Nash, this paper uses the theory of random differential equations and discusses the uncertainty of convergence system caused by various factors. The results show that the results of mean square deviation obtained by the confluence model with random parameters is smaller than that of relative to the convergence model with random input, which is because mean square deviation obtained by confluence model with random parameters mainly depends on variation coefficient of storage coefficient and random storage coefficient changing size directly affects uncertainty degree of flood flow process. The flow process is no longer a definite numerical, which obtained by the confluence model by considering the influence of random factors, but it´s a probability form complying with a certain distribution.