Multi-objective optimization of rainfall-runoff models using the evolution strategy
2004
Fujihara, Y. (Kobe Univ. (Japan)) | Tanakamaru, H. | Hata, T. | Tada, A.
abstract In this study, we apply the Evolution Strategy (ES), which is one of the global optimization methods, to multi-objective optimization of the Tank Model parameters. The main feature of the ES is that its principal search procedure is mutation, which is conducted by adding perturbation of normal random variable to decision variable, and the size of perturbation is updated by self-adaptation. In the algorithm of the multiobjective ES, the Pareto ranking is adopted as a fitness evaluation. RMSE (root mean square error) which emphasizes the error at high flows and RR (root mean square of relative error) which emphasizes the errorat low flows are used as objective functions. Firstly, 21 strict Pareto-optimal solutions are obtained using the weighting method based on a single-objective search. And then, about 200 Pareto-optimal solutions are obtained using the multi-objective ES. Pareto-optimal solutions by the ES are almost the same as solutions by the weighting method and a large number of Pareto-optimal solutions are obtained by relatively small amount of calculation. It is shown that the ES is effective and efficient in multi-objective optimization of rainfall-runoff models.
显示更多 [+] 显示较少 [-]