Training Sample Selection Strategy Applied to CNN in Magneto-Thermal Coupled Analysis
- 9 February 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 57 (6), 1-4
- https://doi.org/10.1109/tmag.2021.3058131
Abstract
Deep learning (DL) has attracted more and more attention in computational electromagnetism. Particularly, the Convolutional Neural Network (CNN) is one of the most popular learning models in DL due to its excellent capacity for feature extraction and convergence. The efficiency of CNN mainly depends on how many training samples are needed to effectively converge the network. The sample preparation process often involves a lot of numerical computations, which can be very expensive and time-consuming. In this paper, based on the traditional DL network training procedure, two different approaches, namely adding smart training samples and reference samples, are proposed to help the DL network converge. The smart sample selection is based on a greedy algorithm, which can be applied for both training and reference samples. The influences of these two approaches on the CNN training process are investigated by an example of the coupled magneto-thermal computation applied to a transformer. Numerical results show that the two proposed approaches can significantly help the network to converge and improve the efficiency of the DL model.Keywords
This publication has 9 references indexed in Scilit:
- Investigation of convolutional neural network U-net under small datasets in transformer magneto-thermal coupled analysisCOMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, 2020
- Topology Optimization Accelerated by Deep LearningIEEE Transactions on Magnetics, 2019
- Deep Learning for Magnetic Field EstimationIEEE Transactions on Magnetics, 2019
- Sensor placement in nuclear reactors based on the generalized empirical interpolation methodJournal of Computational Physics, 2018
- Analysis of Eddy Current Effect and Loss Calculation of Transformer Winding Based on Finite Element AlgorithmPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- A survey of deep neural network architectures and their applicationsNeurocomputing, 2016
- A Deep Learning-Based Approach to Progressive Vehicle Re-identification for Urban SurveillancePublished by Springer Science and Business Media LLC ,2016
- A prioriconvergence of the Greedy algorithm for the parametrized reduced basis methodESAIM: Mathematical Modelling and Numerical Analysis, 2012
- New development in freefem++Journal of Numerical Mathematics, 2012