Multi-objective Antenna Optimization Research Articles

Multistage Collaborative Machine Learning and its Application to Antenna Modeling and Optimization

A multistage collaborative machine learning (MS-CoML) method that can be applied to efficient multiobjective antenna modeling and optimization is proposed. Machine learning methods, including single-output Gaussian process regression (SOGPR) and symmetric and asymmetric multioutput GPR (MOGPR) methods, are introduced to collaboratively build highly accurate multitask surrogate models for antennas. Variable-fidelity electromagnetic (EM) models are simulated, with their responses utilized to build separate MOGPR surrogate models. By combining the three machine-learning methods in a multistage framework, mappings between the same and different responses of the EM models with variable fidelity are learned, therein helping to substantially reduce the computational effort under a negligible loss of predictive power. Three antenna designs aiming at single-band, broadband, and multiband applications are selected as examples. And, for illustrating the applicability and superiority of the proposed MS-CoML method, a reference point-based multiobjective antenna optimization algorithm is used to optimize these three antennas. Simulation results show that using the MS-CoML method can significantly reduce the total optimization time without compromising modeling accuracy and optimized performance.

Rapid multi-objective optimization of antennas using nested kriging surrogates and single-fidelity EM simulation models

PurposeThis study aims to propose a computationally efficient framework for multi-objective optimization (MO) of antennas involving nested kriging modeling technology. The technique is demonstrated through a two-objective optimization of a planar Yagi antenna and three-objective design of a compact wideband antenna.Design/methodology/approachThe keystone of the proposed approach is the usage of recently introduced nested kriging modeling for identifying the design space region containing the Pareto front and constructing fast surrogate model for the MO algorithm. Surrogate-assisted design refinement is applied to improve the accuracy of Pareto set determination. Consequently, the Pareto set is obtained cost-efficiently, even though the optimization process uses solely high-fidelity electromagnetic (EM) analysis.FindingsThe optimization cost is dramatically reduced for the proposed framework as compared to other state-of-the-art frameworks. The initial Pareto set is identified more precisely (its span is wider and of better quality), which is a result of a considerably smaller domain of the nested kriging model and better predictive power of the surrogate.Research limitations/implicationsThe proposed technique can be generalized to accommodate low- and high-fidelity EM simulations in a straightforward manner. The future work will incorporate variable-fidelity simulations to further reduce the cost of the training data acquisition.Originality/valueThe fast MO optimization procedure with the use of the nested kriging modeling technology for approximation of the Pareto set has been proposed and its superiority over state-of-the-art surrogate-assisted procedures has been proved. To the best of the authors’ knowledge, this approach to multi-objective antenna optimization is novel and enables obtaining optimal designs cost-effectively even in relatively high-dimensional spaces (considering typical antenna design setups) within wide parameter ranges.

Fast Multi-Objective Antenna Optimization Based on RBF Neural Network Surrogate Model Optimized by Improved PSO Algorithm

In this paper, a radial basis function neural network (RBFNN) surrogate model optimized by an improved particle swarm optimization (PSO) algorithm is developed to reduce the computation cost of traditional antenna design methods which rely on high-fidelity electromagnetic (EM) simulations. Considering parameters adjustment and update mechanism simultaneously, two modifications are proposed in this improved PSO. First, time-varying learning factors are designed to balance exploration and exploitation ability of particles in the search space. Second, the local best information is added to the updating process of particles except for personal and global best information for better population diversity. The improved PSO is applied to train RBFNN for determining optimal network parameters. As a result, the constructed improved PSO-RBFNN model can be used as a surrogate model for antenna performance prediction with better network generalization capability. By integrating the improved PSO-RBFNN surrogate model with multi-objective evolutionary algorithms (MOEAs), a fast multi-objective antenna optimization framework for multi-parameter antenna structures is then established. Finally, a Pareto-optimal planar miniaturized multiband antenna design is presented, demonstrating that the proposed model provides better prediction performance and considerable computational savings compared to those previously published approaches.

Fast Multi-Objective Optimization of Multi-Parameter Antenna Structures Based on Improved BPNN Surrogate Model

In this paper, a surrogate model based on a sparsely connected back propagation neural networks (SC-BPNN) is proposed to reduce the large computational cost of conventional multi-objective antenna optimization problems. In this model, the connection parameters and network structure can be adaptively tuned by a hybrid real-binary particle swarm optimization (HPSO) algorithm for better network global optimization capability. Also, a time-varying transfer function is introduced to improve the problem of easily trapping into local optimum and to accelerate network convergence. Further, a fast multi-objective optimization framework based on the proposed SC-BPNN is established for multi-parameter antenna structures. Finally, a Pareto-optimal planar miniaturized multiband antenna design is presented, indicating that the proposed model predicts antenna performance more accurately and saves considerable computational cost compared to those previously published approaches.

Patch size setup and performance/cost trade-offs in multi-objective EM-driven antenna optimization using sequential domain patching

PurposeThis paper aims to assess control parameter setup and its effect on computational cost and performance of deterministic procedures for multi-objective design optimization of expensive simulation models of antenna structures.Design/methodology/approachA deterministic algorithm for cost-efficient multi-objective optimization of antenna structures has been assessed. The algorithm constructs a patch connecting extreme Pareto-optimal designs (obtained by means of separate single-objective optimization runs). Its performance (both cost- and quality-wise) depends on the dimensions of the so-called patch, an elementary region being relocated in the course of the optimization process. The cost/performance trade-offs are studied using two examples of ultra-wideband antenna structures and the optimization results are compared to draw conclusions concerning the algorithm robustness and determine the most advantageous control parameter setups.FindingsThe obtained results indicate that the investigated algorithm is very robust, i.e. its performance is weakly dependent on the control parameters setup. At the same time, it is found that the most suitable setups are those that ensure low computational cost, specifically non-uniform ones generated on the basis of sensitivity analysis.Research limitations/implicationsThe study provides recommendations for control parameter setup of deterministic multi-objective optimization procedure for computationally efficient design of antenna structures. This is the first study of this kind for this particular design procedure, which confirms its robustness and determines the most suitable arrangement of the control parameters. Consequently, the presented results permit full automation of the surrogate-assisted multi-objective antenna optimization process while ensuring its lowest possible computational cost.Originality/valueThe work is the first comprehensive validation of the sequential domain patching algorithm under various scenarios of its control parameter setup. The considered design procedure along with the recommended parameter arrangement is a robust and computationally efficient tool for fully automated multi-objective optimization of expensive simulation models of contemporary antenna structures.

GUI design exploration software for microwave antennas

Abstract Optimizers in commercial electromagnetic (EM) simulation software packages are the main tools for performing antenna design exploration today. However, these general purpose optimizers are facing challenges in optimization efficiency, supported optimization types and usability for antenna experts without deep knowledge on optimization. Aiming to fill the gaps, a new antenna design exploration tool, called Antenna Design Explorer (ADE), is presented in this paper. The key features are: (1) State-of-the-art antenna design exploration methods are selected and embedded, addressing efficient antenna optimization (critical but unable to be solved by existing tools) and multiobjective antenna optimization (not available in most existing tools); (2) Human-computer interaction for the targeted problem is studied, addressing various usability issues for antenna design engineers, such as automatic algorithmic parameter setting and interactive stopping criteria; (3) Compatibility with existing tools is studied and ADE is able to co-work with existing EM simulators and optimizers, combining advantages. A case study verifies the advantages of ADE. Highlights A new antenna design exploration tool, called Antenna Design Explorer (ADE), is presented in this paper. State-of-the-art antenna design exploration methods are selected and embedded, addressing efficient antenna optimization (critical but difficult to be solved by existing tools) and multiobjective antenna optimization (not available in most existing tools). Human-computer interaction for the targeted problem is studied, addressing various usability issues for antenna design engineers. Compatibility with existing tools is studied and ADE is able to co-work with existing EM simulators and optimizers, combining advantages.

Efficient Multi-Objective Simulation-Driven Antenna Design Using Co-Kriging

A methodology for fast multi-objective antenna optimization is presented. Our approach is based on response surface approximation (RSA) modeling and variable-fidelity electromagnetic (EM) simulations. In the design process, a computationally cheap RSA surrogate model constructed from sampled coarse-discretization EM antenna simulations is optimized using a multi-objective evolutionary algorithm. The initially determined Pareto optimal set representing the best possible trade-offs between conflicting design objectives is then iteratively refined. In each iteration, a limited number of high-fidelity EM model responses are incorporated into the RSA model using co-kriging. The enhanced RSA model is subsequently re-optimized to yield the refined Pareto set. Combination of low- and high-fidelity simulations as well as co-kriging results in the low overall optimization cost. The proposed approach is validated using two UWB antenna examples.

Low-cost EM-simulation-driven Multi-objective Optimization of Antennas

A surrogate-based method for efficient multi-objective antenna optimization is presented. Our technique exploits response surface approximation (RSA) model constructed from sampled low-fidelity antenna model (here, obtained through coarse-discretization EM simulation). The RSA model enables fast determination of the best available trade-offs between conflicting design goals. A low-cost RSA model construction is possible through initial reduction of the design space. Optimization of the RSA model has been carried out using a multi-objective evolutionary algorithm (MOEA). Additional response correction techniques have been subsequently applied to improve selected designs at the high-fidelity EM antenna model level. The refined designs constitute the final Pareto set representation. The proposed approach has been validated using an ultra-wideband (UWB) monocone and a planar Yagi-Uda antenna.