An ASGC-Surrogate Model Based Yield Optimization for Passive Microwave Devices with Differential Evolution Algorithm

The performance of microwave devices is affected by many uncertainties during the manufacturing process. Proper design with least influence from these disturbances is required for optimal yield. At the design stage of microwave devices, the electromagnetic (EM) based simulation methods are accurate and comprehensive, but it suffers from expensive computational cost. For yield optimization, highly efficient optimization algorithms are necessary. In this article, a global surrogate model is constructed on the basis of the adaptive hierarchical sparse grid collocation (ASGC) algorithm to significantly reduce the number of sampling points. The differential evolution algorithm combined with the ASGC-based surrogate model is proposed to decrease the total simulation time. Finally, to validate the accuracy and efficiency of the proposed algorithm, several representative examples are presented.