An Adaptive Interpolating MLS Based Response Surface Model Applied to Design Optimizations of Electromagnetic Devices

A response surface model (RSM) based on a combination of interpolating moving least squares and multistep method is proposed. The proposed RSM can automatically adjust the supports of its weight functions according to the distribution of the sampling points when it is used to reconstruct a computationally heavy design problem. Numerical examples are given to demonstrate the feasibility and efficiency of the proposed method for solving inverse problems