Multi-step Optimization of Parameters in the Xinanjiang Model Taking Into Account Their Time Scale Dependency

Abstract

The objective of this study is to find an efficient way to optimize parameters in the Xinanjiang model automatically. According to the time scale dependent sensitivities, a multi-step optimization scheme is developed with the help of the SCEM-UA algorithm. Some case studies were designed to test the efficiency of the scheme. The tests show this scheme under ideal condition could converge to the true values of parameters to be optimized. For the actual practice, the multi-step optimization scheme is capable of searching the optimum parameter set for the Xinanjiang model, and the simulation results are acceptable.