In recent years, we have done two tasks in the optimization of the algorithm by using the genetic algorithm to set the parameters of the Xin ' an river model. One is to introduce the simulated annealing algorithm to constrain the mutation operator, namely the genetic simulated annealing algorithm, and the second is the real adaptive genetic algorithm. Details such as the following:
watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvz2lzzxjfd2h1/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/ Dissolve/70/gravity/center ">
The number of parameters is determined as shown in the following two images.
watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvz2lzzxjfd2h1/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/ Dissolve/70/gravity/center ">
It is necessary to state that these two graphs are the result of determining the percentage of data in the same year of the watershed. Visible, there is a clear "different participation and efficiency" phenomenon.
The next step is to make uncertainty analysis of the model parameters in order to better rate the model parameters.
Optimization strategy of genetic algorithm