Comparison of response surface method and Kriging method for approximation modeling
Xiaobo Zhang · 2017 2nd International Conference on Power and Renewable Energy (ICPRE) · 2017
For getting the efficient approximation modeling method. Two approximation modeling methods are discussed in this paper. Three test functions are used for estimating the approximation model. In order to get the computational accuracy and computational efficiency the approximation models for both RSM and Kriging method are established. By employing the approximation models the performance are obtained and the results are given. From the result, it can be seen that the Kriging method has high computational accuracy, while the computational time is much larger than the RSM. The RSM is more suitable for low order nonlinear problem. And Kriging method is suitable for high order nonlinear problem. If the problem no requirement on computational time, Kriging method can be used for its high in computational accuracy.