Open AccessData‐driven model based design and analysis of antenna structures
Author(s) -
Ulaganathan Selvakumar,
Koziel Slawomir,
Bekasiewicz Adrian,
Couckuyt Ivo,
Laermans Eric,
Dhaene Tom
Publication year - 2016
Publication title -
iet microwaves, antennas and propagation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.555
H-Index - 69
eISSN - 1751-8733
pISSN - 1751-8725
DOI - 10.1049/iet-map.2016.0089
Subject(s) - kriging , curse of dimensionality , computation , computer science , antenna (radio) , wideband , gaussian process , sensitivity (control systems) , algorithm , function (biology) , gaussian , mathematical optimization , electronic engineering , mathematics , engineering , machine learning , telecommunications , evolutionary biology , biology , physics , quantum mechanics
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 computation‐intensive 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 (GEK) , is proposed in this work to achieve cost‐efficient modelling of antenna structures. The GEK 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 the less number of training samples over the Ordinary Kriging (OK) approach which utilises function data only. The discussed technique has been favourably compared with OK in terms of computational cost.