Multifrequency nonlinear model of magnetic material with artificial intelligence optimization

Pawłowski, Jarosław; Kutorasiński, K.; Szewczyk, Marcin

doi:10.1038/s41598-022-23810-9

“…When the performance is evaluated using finite element (FE) models, this becomes prohibitively expensive, especially when saturation current is considered. Adapting of surrogate models (lizadeh et al , 2020) can obviously reduce the optimization time, and neural network (NN) (Sasaki and Igarashi, 2019; Pawłowski et al , 2022) is one of the commonly used and powerful surrogate models. However, the suitability of using surrogate models for specific scenarios requires discussion: for example, the training time for surrogate models containing complex features may be substantially longer than the time for a single conventional optimization.…”

Section: Introductionmentioning

confidence: 99%

Multi-objective optimization and uncertainty quantification for inductors based on neural network

Kong,

Yin,

Gong

et al. 2024

COMPEL

0

View full text Add to dashboard Cite

Purpose The prolonged training time of the neural network (NN) has sparked considerable debate regarding their application in the field of optimization. The purpose of this paper is to explore the beneficial assistance of NN-based alternative models in inductance design, with a particular focus on multi-objective optimization and uncertainty analysis processes. Design/methodology/approach Under Gaussian-distributed manufacturing errors, this study predicts error intervals for Pareto points and select robust solutions with minimal error margins. Furthermore, this study establishes correlations between manufacturing errors and inductance value discrepancies, offering a practical means of determining permissible manufacturing errors tailored to varying accuracy requirements. Findings The NN-assisted methods are demonstrated to offer a substantial time advantage in multi-objective optimization compared to conventional approaches, particularly in scenarios where the trained NN is repeatedly used. Also, NN models allow for extensive data-driven uncertainty quantification, which is challenging for traditional methods. Originality/value Three objectives including saturation current are considered in the multi-optimization, and the time advantages of the NN are thoroughly discussed by comparing scenarios involving single optimization, multiple optimizations, bi-objective optimization and tri-objective optimization. This study proposes direct error interval prediction on the Pareto front, using extensive data to predict the response of the Pareto front to random errors following a Gaussian distribution. This approach circumvents the compromises inherent in constrained robust optimization for inductance design and allows for a direct assessment of robustness that can be applied to account for manufacturing errors with complex distributions.

show abstract

“…This study explores the effectiveness of classical, non-AI-based algorithms in developing a time-dependent model for magnetic rings that operate across a wide frequency range and under nonlinear conditions 8,9 . The model relies on impedance measurements 10 that are frequency-dependent and obtained under a DC-bias current 11 .…”

Section: Introduction Problem Definition and Goalmentioning

confidence: 99%

Nonlinear Modeling of Magnetic Materials for Circuit Simulations

Kutorasiński

¹

,

Pawłowski

²

,

Szewczyk

³

et al. 2023

Preprint

0

View full text Add to dashboard Cite

Magnetic materials in the form of magnetic rings are widely used in power engineering products. In many cases, they operate in high frequency and in nonlinear conditions, e.g., as damping elements in electrical power substations equipped with Gas-Insulated Switchgear (GIS) where they provide suppression of Very Fast Transient Overvoltages (VFTOs). To model phenomena in GIS with magnetic rings it is required to have realistic time-dependent models of magnetic materials operating in a wide frequency range and nonlinear conditions. Nowadays, this has become even more relevant due to the actual trend in the industry to create digital twins of physical devices. Models composed of high-precise discrete lumped nonlinear elements are in demand in circuit simulators like SPICE. This work introduces a method based on classical algorithms that find equivalent lumped models of magnetic cores based on frequency-dependent measurements of impedance under DC-bias current. The model is specifically designed to achieve numerically stable simulations.

show abstract

“…This study explores the effectiveness of classical, non-AI-based algorithms in developing a time-dependent model for magnetic rings that operate across a wide frequency range and under nonlinear conditions 8 , 9 . The model relies on impedance measurements 10 that are frequency-dependent and obtained under a DC-bias current 11 .…”

Section: Introductionmentioning

confidence: 99%

Nonlinear modeling of magnetic materials for circuit simulations

Kutorasiński,

Pawłowski,

Leszczyński

et al. 2023

Sci Rep

1

0

View full text Add to dashboard Cite

Magnetic materials in the form of magnetic rings are widely used in power engineering products. In many cases, they operate in high frequency and in nonlinear conditions, e.g., as damping elements in electrical power substations equipped with Gas-Insulated Switchgear (GIS) where they provide suppression of Very Fast Transient Overvoltages (VFTOs). To model phenomena in GIS with magnetic rings it is required to have realistic time-dependent models of magnetic materials operating in a wide frequency range and nonlinear conditions. Nowadays, this has become even more relevant due to the actual trend in the industry to create digital twins of physical devices. Models composed of high-precise discrete lumped nonlinear elements are in demand in circuit simulators like SPICE. This work introduces a method based on classical algorithms that find equivalent lumped models of magnetic cores based on frequency-dependent measurements of impedance under DC-bias current. The model is specifically designed to have smooth behavior in the current domain and thanks to that to improve numerical stability in the time domain simulations.

show abstract

Multifrequency nonlinear model of magnetic material with artificial intelligence optimization

Cited by 5 publications

References 30 publications

Multi-objective optimization and uncertainty quantification for inductors based on neural network

Multi-objective optimization and uncertainty quantification for inductors based on neural network

Nonlinear Modeling of Magnetic Materials for Circuit Simulations

Nonlinear modeling of magnetic materials for circuit simulations

Contact Info

Product

Resources

About