A hybrid aggressive space-mapping algorithm for EM optimization

We propose a novel hybrid aggressive space-mapping (HASM) optimization algo rithm. HASM exploits both the trust-region aggressive space-mapping (TRASM) strategy and direct optimization, Severe differences between the coarse an d fine models and nonuniqueness of the parameter extraction procedure may c ause the TRASM algorithm to be trapped in local minima. The HASM algorithm is based on a novel lemma that enables smooth switching from the TRASM opti mization to direct optimization if the TRASM algorithm is not converging, I t also enables switching back from direct optimization to the TRASM algorit hm in a smooth may, The uniqueness of the extraction step is improved by ut ilizing a good starting point, The algorithm does not assume that the final space-mapped design is the true optimal design and is robust against sever e misalignment between the coarse and fine models, The examples include a s even-section waveguide transformer, the design of an H-plane waveguide filt er, and a double-folded stub filter.