A Space-Mapping Approach to Microwave Device Modeling Exploiting Fuzzy Systems

IET Microwaves, Antennas & Propagation

Bekasiewicz

et al. 2016

Self Cite

Data-driven models, or metamodels, offer an efficient way to mimic the behaviour of computation-intensive simulators. Subsequently, the usage of such computationally cheap metamodels is indispensable in the design of contemporary antenna structures where computationintensive simulations are often performed in a large scale. Although metamodels offer sufficient flexibility and speed, they often suffer from an exponential growth of required training samples as the dimensionality of the problem increases. In order to alleviate this issue, a Gaussian process based approach, known as Gradient Enhanced Kriging, is proposed in this work to achieve cost-efficient modeling of antenna structures. The Gradient Enhanced Kriging approach incorporates adjoint-based sensitivity data in addition to function data obtained from electromagnetic simulations. The approach is illustrated using a dielectric resonator and an ultra-wideband antenna structures. The method demonstrates significant accuracy improvement with less number of training samples over the Ordinary Kriging approach which utilises function data only. The discussed technique has been favorably compared with Ordinary Kriging in terms of computational cost.

Section: Introductionmentioning

confidence: 99%

Data‐driven model based design and analysis of antenna structures

Ulaganathan

IET Microwaves, Antennas & Propagation

Bekasiewicz

et al. 2016

Self Cite

Section: Introductionmentioning

confidence: 99%

Cost-efficient modeling of antenna structures using Gradient-Enhanced Kriging

Ulaganathan

2015 Loughborough Antennas &Amp; Propagation Conference (LAPC)

Bekasiewicz

et al. 2015

Self Cite

Abstract-Reliable yet fast surrogate models are indispensable in the design of contemporary antenna structures. Data-driven models, e.g., based on Gaussian Processes or support-vector regression, offer sufficient flexibility and speed, however, their setup cost is large and grows very quickly with the dimensionality of the design space. In this paper, we propose cost-efficient modeling of antenna structures using GradientEnhanced Kriging. In our approach, the training data set contains, apart from the EM-simulation responses of the structure at hand, also derivative data at the respective training locations obtained at little extra cost using adjoint sensitivity techniques. We demonstrate that introduction of the derivative information into the model allows for considerable reduction of the model setup cost (in terms of the number of training points required) without compromising its predictive power. The Gradient-Enhanced Kriging technique is illustrated using a dielectric resonator antenna structure. Comparison with conventional Kriging interpolation is also provided.

“…SM modeling with variable weight coefficients [25] provides efficient utilization of available fine model data, however, at the expense of computational overhead related to a separate parameter extraction process required for each evaluation of the surrogate. SM modeling enhanced by fuzzy systems [26], radial basis functions [22], or Kriging [27] offer accuracy comparable with [25] without compromising computational cost; however, the implementation of these models is somewhat complicated.…”

Section: Introductionmentioning

confidence: 99%

Variable-Fidelity Electromagnetic Simulations and Co-Kriging for Accurate Modeling of Antennas

IEEE Trans. Antennas Propagat.

Ogurtsov

Couckuyt

et al. 2013

Self Cite

Abstract-Accurate and fast models are indispensable in contemporary antenna design. In this paper, we describe the low-cost antenna modeling methodology involving variable-fidelity electromagnetic (EM) simulations and co-Kriging. Our approach exploits sparsely sampled accurate (high-fidelity) EM data as well as densely sampled coarse-discretization (low-fidelity) EM simulations that are accommodated into one model using the co-Kriging technique. By using coarse-discretization simulations, the computational cost of creating the antenna model is greatly reduced compared to conventional approaches, where high-fidelity simulations are directly used to set up the model. At the same time, the modeling accuracy is not compromised. The proposed technique is demonstrated using three examples of antenna structures. Comparisons with conventional modeling based on high-fidelity data approximation, as well as applications for antenna design, are also discussed.