Reliable Simulation-Driven Design Optimization of Microwave Structures Using Manifold Mapping

Abstract

Abstract-A computationally efficient surrogate-based framework for reliable simulation-driven design optimization of microwave structures is described. The key component of our algorithm is manifold mapping, a response correction technique that aligns the coarse model (computationally cheap representation of the structure under consideration) with the accurate but CPU-intensive (fine) model of the optimized device. The parameters of the manifold mapping surrogate are explicitly calculated based on the fine model data accumulated during the optimization process. Also, manifold mapping does not use any extractable parameters, which makes it easy to implement. Robustness and excellent convergence properties of the proposed algorithm are demonstrated through the design of several microwave devices including microstrip filters and a planar antenna.

Obstructive uropathy caused by an inguinal hernia in a kidney transplant recipient: Report of hernia cure by the shouldice technique

“…Other efficient surrogate‐based techniques include manifold mapping [26] recently applied in microwave design [27], adaptive response correction [28], as well as shape‐preserving response prediction [29]. These approaches have certain advantage over SM in the sense that they do not require parameter extraction process (essential for SM) that might be CPU intensive; however, the low‐fidelity model is still recommended to be substantially cheaper than the high‐fidelity one.…”

A Multi-Fidelity Based Adaptive Sampling Optimisation Approach for the Rapid Design of Double-Negative Metamaterials

“…An optimisation design space with widely set bounds can be searched effectively using local surrogate based methodologies, such as space and manifold mapping [5,6] if the design space is unimodal. However, it must be assumed that there may be local basins of attraction, with the prospect that the objective function multi-dimensional landscape is highly nonlinear.…”

Aerodynamic shape optimization by multi-fidelity modeling and manifold mapping

“…Over the last few decades, various correction techniques and related optimization algorithms have been developed, including the approximation and model management optimization (AMMO) [30], multi-point correction techniques [33,34], several variations of output space mapping (SM) [33], as well as manifold mapping (MM) [1,35]. Apart from the aforementioned ones, which are all so-called parametric methods [31] (where the correction functions are given explicitly with the parameters usually obtained by explicit calculations or solving auxiliary linear regression problems), a number of nonparametric technique have been developed, such as the shape-preserving response prediction (SPRP) [36], adaptive response correction (ARC) [37], and adaptive response prediction (ARP) [38].…”

“…In this chapter, the direct and multi-fidelity optimization algorithms are applied to two benchmark aerodynamic design problems involving inviscid and viscous transonic flow past airfoil shapes. These benchmark cases were developed by the AIAA Aerody- [63,35]. Moreover, for comparison purposes, we solve both benchmark cases with a gradient-based technique with adjoints and trust regions [21].…”

“…The surrogate model s is a suitably corrected low-fidelity model c. The key component of the multi-fidelity optimization algorithm is the physics-based low-fidelity (or coarse) model c that embeds certain knowledge about the system under consideration, and allows us to construct a reliable surrogate using a limited amount of high-fidelity model data. In this work, the low-fidelity model is evaluated using the same CFD solver as the high-fidelity model f. Two (parametric) correction methods are considered: SM[34] (for the sake of comparison and validation) and manifold mapping (MM)[63,35].…”