Abstract:Geometry scaling of microwave circuits is an essential but challenging task. In particular, the employment of a given passive structure in a different application area often requires re-adjustment of the operating frequencies/bands while maintaining top performance. Achieving this necessitates the utilization of numerical optimization methods. Nonetheless, if the intended frequencies are distant from the ones at the starting point, local search procedures tend to fail, whereas global search algorithms are computationally expensive. As recently demonstrated, a combination of large-scale concurrent geometry parameter scaling with intermittent local tuning allows for dependable re-design of high-frequency circuits at low CPU costs. Unfortunately, the procedure is only applicable to single-band structures due to synchronized modifications of all operating bands under scaling. This article discusses a novel procedure that leverages a similar overall concept, but allows for independent control of all center frequencies. To achieve this goal, an automated decision-making procedure is developed in which a set of orthogonal scaling directions are determined based on their effect on individual circuit bands, and using auxiliary optimization sub-problems. The scaling range is then automatically computed by solving an appropriately-defined least-square design relocation problem. The methodology introduced in the work is illustrated using two planar passive devices. In both cases, wide-range operating frequency re-design has been demonstrated and favorably compared to conventional gradient-based tuning. Furthermore, the presented procedure has been shown to be computationally efficient. It is also easy to implement and integrate with a variety of gradient-based optimization procedures of a descent type.

Adaptively Constrained Parameter Extraction for Robust Space Mapping Optimization of Microwave Circuits

Simulation-Based Robust Design of Complex Product Considering Uncertainties of Metamodel, Design Variables, and Noise Parameters

Robust 3-D Shape Optimization of Electromagnetic Devices by Combining Sensitivity Analysis and Adaptive Geometric Parameterization

Leveraging Parallel Speedup for Intelligent Design of Metasurface

Machine-learning-based global optimization of microwave passives with variable-fidelity EM models and response features

Comparative Study of Surrogate Modeling Methods for Signal Integrity and Microwave Circuit Applications

Parallel Space-Mapping Approach to EM Optimization

A Novel Surrogate-Based Approach to Yield Estimation and Optimization of Microwave Structures Using Combined Quadratic Mappings and Matrix Transfer Functions

Local-global space mapping for rapid EM-driven design of compact RF structures

Tabu search based robust design optimization for inverse problems in electromagnetics

Developments of an Efficient Global Optimal Design Technique – a Combined Approach of MLS and SA Algorithm

Robust Optimal Design of Low Frequency Metamaterials Considering the Fabrication Imperfections and Parameter Fluctuations

An Efficient Robust Optimization Methodology for Metamaterial Unit Design

Efficient EM Optimization Exploiting Parallel Local Sampling Strategy and Bayesian Optimization for Microwave Applications

Conformal Antenna Optimization with Space Mapping

Rapid Surrogate-Aided Multicriterial Optimization of Compact Microwave Passives Employing Machine Learning and ANNs

Expedited re-design of multi-band passive microwave circuits using orthogonal scaling directions and gradient-based tuning

An Improved Multi-Objective GA for Low Frequency Metamaterial Unit Robust Optimization under Uncertainty

EM-Centric Multiphysics Optimization of Microwave Components Using Parallel Computational Approach

Advanced Mesh Space Mapping Approach With Fast Coarse Mesh Models Comprising Sharpening Structural Processing and Mesh Deformation

Machine Learning based parameter tuning strategy for MMC based topology optimization