Selection of genetic algorithm operators and application for calibration of dissolved oxygen simulation in the river water quality model

In the overall model development, model calibration is one of the most important tasks. This paper focuses on use of genetic algorithm (GA) in water quality model calibrations and definition of suitable operators which allow GA process run well and generates sets of optimized model parameters for river water quality model. An investigation on the importance of GA operators on model parameter optimization was considered in this study based on the data which collected from the Tatara River, Fukuoka, Japan. Many of numerical experiments on GA operators were conducted to check impacts of different GA operators on efficiency of GA process and goodness of parameter set which will be used for calibrations of water quality model. It was found that most of GA operators have sensitive effects on optimization process, in which the most important operators are rate of crossover and mutation.