Search

Electromagnetic (EM) optimization is an essential part of an EM design cycle. In this paper, for the first time, a decomposition method is formulated to address the challenges of EM optimization with many design variables. Single large EM optimization is decomposed into multiple smaller suboptimizations to improve the optimization efficiency. In practical EM optimizations, there are no two design variables whose effect on the output response is completely independent. We propose a decomposition method, which uses the second-order derivative information to measure the level of dependence (measure of interaction) between the design variables and thereby identify, separate, and group the design variables to form multiple smaller suboptimizations. Each suboptimization includes generating multiple fine model responses for constructing the surrogate model and its optimization using trust region. Multiple suboptimizations are formulated to be independent, so that parallel computations can be exploited to evaluate concurrent suboptimization updates. The resultant vector, a combined vector of suboptimization updates, will be closer to the optimal solution. Furthermore, the step size of the optimization update in the proposed approach is much larger in comparison with the step size of the optimization update without decomposition. Using this approach, the overall large optimization update decreases the value of the objective function more efficiently and quickly. Two examples of EM optimization are used to illustrate the proposed technique. [ABSTRACT FROM PUBLISHER]

Conventional EM optimization aims to use fewest possible fine model evaluations to increase the speed of optimization. In this work, we propose to use a large number of fine model evaluations to achieve an overall speedup. A large number of fine model evaluations allows us to build a surrogate model valid in a large neighborhood. In the proposed technique, these valid surrogate models are used to achieve large and effective optimization updates, thereby resulting in fewer iterations of the optimization process. Valid surrogate models uses many fine model evaluations which are realized in parallel using hybrid distributed shared memory computing platforms. Parallel computation of large number of fine model evaluations reduces the major computational time required for constructing a surrogate model. Furthermore, we exploit trust region algorithms to guarantee convergence and to re-define the fine model evaluation range in each iteration of the proposed optimization algorithm. The proposed technique aims to increase the speed of gradient based EM optimization when no coarse model (e.g., empirical or equivalent circuits) is available. Three typical examples are used to illustrate the proposed technique. [ABSTRACT FROM AUTHOR]