Find in Library

There are a considerable number of accuracy indexes for surrogate models to measure. However, most of these indexes are not capable of comparing surrogate models of different data types. The drawbacks of the existing indexes are firstly discussed. Then a new index named “Mean Normalized Absolute Error(MNEA)” is introduced. Kriging models of 11 different types of aerodynamic parameters from a tactical missile model are built, with the adding-points design of experiment method and cross-validation verification methods. The effectiveness of MNEA is verified by comparing the MNEA value and the correlation plots of the example, which proves that MNEA can be used to compare the accuracy of surrogate models of different data types.