Three-Level Output Space Mapping Strategy for Electromagnetic Design Optimization

Ayed, Ramzi Ben; Gong, Jinlin; Brisset, S.; Gillon, F.; Brochet, P.

doi:10.1109/tmag.2011.2174349

Cited by 47 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the fine models, such as 3D EM simulator, are typically very accurate but computationally very intensive. By using space mapping technique, we can combine the accuracy of the fine models with the computational efficiency of the coarse models [11], [12]. A neuro-space mapping algorithm which uses the neural networks to build the relationships between the fine models and the coarse models is introduced in [13].…”

Section: Introductionmentioning

confidence: 99%

Multiphysics Parametric Modeling of Microwave Components Using Combined Neural Networks and Transfer Function

et al. 2020

View full text Add to dashboard Cite

This paper proposes a new technique to develop an accurate multiphysics parametric model for microwave components to speed up the multiphysics modeling process. In the proposed technique, the artificial neural networks (ANNs) and pole/residue based transfer function are incorporated to represent the highly non-linear relationships between electromagnetic centric (EM-centric) multiphysics behaviors and multiphysics geometrical/non-geometrical design parameters. Vector fitting technique is utilized to obtain the poles/residues of the transfer function for each multiphysics sample. Since the relationship between multiphysics design parameters and the pole/residues of the transfer function is non-linear and unknown, two mapping functions are proposed to establish the mathematical links between the multiphysics design parameters and poles/residues. Parallel multiphysics data generation is proposed to generate the training and testing data for establishing the proposed multiphysics parametric model. A two stage training algorithm is proposed to guide the multiphysics training process. Once an accurate overall model is developed, it can be used to provide accurate and fast prediction of the multiphysics behavior of microwave components with geometrical and non-geometrical parameters as variables, and further can be used in the high level design. Compared with the existing multiphysics modeling methods, the proposed technique can achieve better model accuracy with high efficiency. The proposed technique provides an accurate and efficient methodology even when the coarse model or empirical model is unavailable. Two microwave examples are used to illustrate the validity of the proposed multiphysics parametric modeling technique. INDEX TERMS Artificial neural networks, multiphysics, parametric modeling, parallel computation, transfer function.

show abstract

Section: Introductionmentioning

confidence: 99%

Multiphysics Parametric Modeling of Microwave Components Using Combined Neural Networks and Transfer Function

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Artificial neural networks (ANNs) have been recognized as powerful tools in microwave modeling and design [1]- [3], such as nonlinear microwave device modeling [4]- [6], parametric modeling [7]- [9], electromagnetic (EM) optimization [10], [11], multiphysics modeling [12], and inverse modeling [13]. ANNs are trained to learn EM/physics data which represent the behavior of microwave devices.…”

Section: Introductionmentioning

confidence: 99%

Automated Neural-Based Modeling of Microwave Devices Using Parallel Computation and Interpolation Approaches

Zhang

Yan

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Automated model generation (AMG) is an automated artificial neural network (ANN) modeling algorithm, which integrates all the subtasks (including adaptive sampling/data generation, model structure adaptation, training, and testing) in neural model development into one unified framework. In existing AMG, most of the time is spent on data sampling and model structure adaptation due to the iterative neural network training and the sequential computation mechanism. In this paper, we propose an advanced AMG algorithm using parallel computation and interpolation approaches to speed up the neural modeling of microwave devices. Efficient interpolation approaches are incorporated to avoid repetitive training of the intermediate neural networks during adaptive sampling process in AMG. Parallel computation formulation based on a multi-processor environment is proposed to further save time during interpolation calculation, data generation, and model structure adaptation process. Examples of automated modeling of two microwave filters are presented to show the advantage of this paper. INDEX TERMS Design automation, modeling, neural networks, parallel computation, interpolation approaches.

show abstract

“…To further improve the efficiency of local optimization, the space mapping technique [9] is widely used. Several important improvements have been made to enhance the reliability and the efficiency of traditional space mapping, such as introducing the human design intuition [10], altering an EM model by embedding suitable tuning elements (port tuning) [11], and the multilevel method [12]. The port tuning method has shown great success in commercial applications for planar filters.…”

Section: Introductionmentioning

confidence: 99%

Global Optimization of Microwave Filters Based on a Surrogate Model-Assisted Evolutionary Algorithm

Liu

Yang

Lancaster

2017

IEEE Trans. Microwave Theory Techn.

View full text Add to dashboard Cite

Local optimization is a routine approach for fullwave optimization of microwave filters. For filter optimization problems with numerous local optima or where the initial design is not near to the optimal region, the success rate of the routine method may not be high. Traditional global optimization techniques have a high success rate for such problems, but are often prohibitively computationally expensive considering the cost of full-wave electromagnetic simulations. To address the above challenge, a new method, called surrogate model-assisted evolutionary algorithm for filter optimization (SMEAFO), is proposed. In SMEAFO, considering the characteristics of filter design landscapes, Gaussian process surrogate modeling, differential evolution operators, and Gaussian local search are organized in a particular way to balance the exploration ability and the surrogate model quality, so as to obtain high-quality results in an efficient manner. The performance of SMEAFO is demonstrated by two real-world design cases (a waveguide filter and a microstrip filter), which do not appear to be solvable by popular local optimization techniques. Experiments show that SMEAFO obtains high-quality designs comparable with global optimization techniques but within a reasonable amount of time. Moreover, SMEAFO is not restricted by certain types of filters or responses. The SMEAFO-based filter design optimization tool can be downloaded from http://fde.cadescenter.com.

show abstract

Three-Level Output Space Mapping Strategy for Electromagnetic Design Optimization

Cited by 47 publications

References 12 publications

Multiphysics Parametric Modeling of Microwave Components Using Combined Neural Networks and Transfer Function

Multiphysics Parametric Modeling of Microwave Components Using Combined Neural Networks and Transfer Function

Automated Neural-Based Modeling of Microwave Devices Using Parallel Computation and Interpolation Approaches

Global Optimization of Microwave Filters Based on a Surrogate Model-Assisted Evolutionary Algorithm

Contact Info

Product

Resources

About