Research On The Parameter Optimization And The Real-time Correction Of Hydrological Model

The parameter estimation and the flood real-time correction are very importantfor the rainfall-runoff hydrologic model. They have a direct relation with improvingthe precision and the interval of flood forecasting. Otherwise, there is an importantmeaning for protecting the property and the life and security of human.In the paper, based on the research results at home and abroad, and according theproblem of traditional hydrological simulation, combining the rainfall-runoffhydrological model and artificial intelligence technique, two problems of theparameter automatic estimation and the real-time correction were investigated, andthe effective models have been established. These models are verified by using somehydrological data of the PanJia-kou reservoir and the results show that the models arevery effectives. The main research contents are as following:(1) Based on the genetic arithmetic, a method on estimating parameter ofrainfall-runoff hydrological model was proposed, that can automatically and quicklyobtain the parameters of hydrological forecast model according to the best simulationof the flood peak discharge and water quantity. The method not only can improveforecast accuracy, but also is an efficient means for the flood forecasting. (2) Utilized autoregressive model of error method. The author designs a real-timecorrection model for flood forecasting of the Panjia-kou reservoir. The correctionresults show that the methods mentioned above are effective. They can better improvethe precision of forecasting and have the advantages of easy to understand.(3) According to the characteristic of flood forecasting, the author designs themanagement and information system of the PanJia-kou reservoir that has amiableinterface and convenient operational mode. Its practicability has been verified in thecase of this paper.Theories and methods developed in the paper are practical, and their outcomes fitfactual situation. Therefore, they can provide valuable reference frame for thereal-time flood forecast of Panjia-kou reservoir.