Skip to Main Content
Practical experience with numerical model calibration suggests that no single objective is adequate to measure the ways in which the model fails to match the important characteristics of the observed data. The multiobjective genetic algorithm (MOGA) is used as automatic calibration method for a wide range of numerical models. The task of estimating the entire Pareto set requires a large number of fitness evaluations in a standard MOGA optimization process. However, it's very time consuming to obtain a value of objective functions in many real-world engineering problems. The NSGA-II-1NN algorithm, an effective and efficient methodology to reduce the number of actual fitness evaluations for solving the multiple-objective global optimization problem, is presented in this paper. The test results for multiobjective calibration show that the proposed method only requires about 38 percent of actual fitness evaluations of the NSGA-II.
Granular Computing, 2005 IEEE International Conference on (Volume:2 )
Date of Conference: 25-27 July 2005