open access

https://hdl.handle.net/2115/92851

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

英語

Uncertainties in electromagnetics

2024-07-30

Emerald Group Publishing

COMPEL : The international journal for computation and mathematics in electrical and electronic

43(4)

890-903

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 article is to explore the beneficial assistance of neural network (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, we predict error intervals for Pareto points and select robust solutions with minimal error margins. Furthermore, we establish 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 neural network is repeatedly utilized. Also, NN models allow for extensive data-driven uncertainty quantification, which is challenging for traditional methods. Originality: 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. We propose direct error interval prediction on the Pareto front, utilizing 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.

学術雑誌論文

Accepted Manuscript

© 2024, Emerald Publishing Limited. This AAM is provided for your own personal use only. It may not be used for resale, reprinting, systematic distribution, emailing, or for any other commercial purpose without the permission of the publisher

0332-1649