PARAMETER IDENTIFICATION TO ESTIMATE THE DISCHARGES OF THE MULTIPLE WATER LEVEL STATIONS BY MULTI-OBJECTIVE OPTIMIZATION

Abstract

 This paper presents the parameter identification of a distributed runoff model applied to the Abe River basin where three water level stations exist. The errors between the observed and calculated discharges at the water level stations are minimized simultaneously by multi-objective optimization algorithms. Comparing the performance of five implementations of the optimization algorithms, two implementations of a genetic algorithm NSGA2 of the statistical software R are better than the others. It turned out to be difficult to represent the whole basin by the parameters optimized to a single water level station. However, it is possible to calculate the discharges which are near the observed ones by selecting the pareto optimal solutions under the limited errors of three water level stations.