Research on Multi-fidelity Surrogate Modeling Method of the problem with non-hierarchical low-fidelity analysis models

Objectives The multi-fidelity surrogate (MFS) modeling technology can reduce the simulation cost during the design process of engineering product. Aim to relax the hierarchical relationship between LF (Low Fidelity) analysis models and broaden the engineering application of MFS, Methods a multi-fidelity surrogate modelling method based on variance-weighted sum (VWS-MFS) for the fusion of multiple non-hierarchical LF data is proposed. Based on non-hierarchical LF data, the proposed method builds LF surrogate models using Kriging technology. By quantifying the uncertainty of the LF surrogate models with variance, the non-hierarchical LF data are weighted to construct a trend function. In addition, the improved hierarchical Kriging (IHK) model is introduced to fuse the HF (High Fidelity) and LF data, enabling the correction coefficient of trend function changing throughout the design space. The proposed method is tested by 9 typical examples. Besides, the application of proposed method on the performance prediction of vibration isolator is carried out. Results According to the experimental results, the proposed method shows higher prediction accuracy than the similar method by more than 85 percent. Besides, the result of the application to vibration isolator shows that the accuracy of proposed method is significantly improved by more than 60 percent compared with the static prediction method. Conclusions The proposed method is able to integrate the HF analysis model and multiple non-hierarchical LF analysis models. While the hierarchical relationship between LF analysis models is relaxed, the information of LF data is mined to the maximum extent.