High-Precision Kriging Modeling Method Based on Hybrid Sampling Criteria

Finding new valuable sampling points and making these better distributed in the design space is key to determining approximate effect of Kriging. To this end, a high-precision Kriging modeling method based on hybrid criteria (HKM-HS) proposed solve problem. In HKM-HS method, two infilling strategies MSE (Mean Square Error) are optimized obtain candidate points. By maximizing (MMSE) model, it can generate first point that likely appear sparse area. avoid ill-conditioned correlation matrix caused by too close distance between any points, MC (MSE Correlation function) criterion formed combining function through multiplication division minimized second point. Furthermore, screening used select final expensive evaluation from Finally, test results sixteen benchmark functions house heating case show effectively enhance accuracy stability contrast with other methods.