An investigation on machine learning predictive accuracy improvement and uncertainty reduction using VAE-based data augmentation

Alsafadi, Farah; Yaseen, Mahmoud; Wu, Xu

doi:10.1016/j.nucengdes.2025.114433

An investigation on machine learning predictive accuracy improvement and uncertainty reduction using VAE-based data augmentation

Journal Article · Thu Sep 11 00:00:00 EDT 2025 · Nuclear Engineering and Design

DOI:https://doi.org/10.1016/j.nucengdes.2025.114433· OSTI ID:3000581

^[1]; ^[1]; ^[1]

North Carolina State University, Raleigh, NC (United States)

The confluence of ultrafast computers with large memory, rapid progress in Machine Learning (ML) algorithms, and the availability of large datasets place multiple engineering fields at the threshold of dramatic progress. However, a unique challenge in nuclear engineering is data scarcity because experimentation on nuclear systems is usually more expensive and time-consuming than most other disciplines. One potential way to resolve the data scarcity issue is deep generative learning, which uses certain ML models to learn the underlying distribution of existing data and generate synthetic samples that resemble the real data. In this way, one can significantly expand the dataset to train more accurate predictive ML models. In this study, our objective is to evaluate the effectiveness of data augmentation using variational autoencoder (VAE)-based deep generative models. We investigated whether the data augmentation leads to improved accuracy in the predictions of a deep neural network (DNN) model trained using the augmented data. Additionally, the DNN prediction uncertainties are quantified using Bayesian Neural Networks (BNN) and conformal prediction (CP) to assess the impact on predictive uncertainty reduction. To test the proposed methodology, we used TRACE simulations of steady-state void fraction data based on the NUPEC Boiling Water Reactor Full-size Fine-mesh Bundle Test (BFBT) benchmark. Here, we found that augmenting the training dataset using VAEs has improved the DNN model’s predictive accuracy, improved the prediction confidence intervals, and reduced the prediction uncertainties.

View Accepted Manuscript (DOE)

Research Organization:: Idaho Operations Office, Idaho Falls, ID (United States); North Carolina State University, Raleigh, NC (United States)

Sponsoring Organization:: USDOE Office of Nuclear Energy (NE)

Grant/Contract Number:: NE0009467

OSTI ID:: 3000581

Journal Information:: Nuclear Engineering and Design, Journal Name: Nuclear Engineering and Design Vol. 445; ISSN 0029-5493

Publisher:: Elsevier BVCopyright Statement

Country of Publication:: United States

Language:: English

References (16)

Bayesian Neural Networks: An Introduction and Survey Goan, Ethan; Fookes, Clinton Case Studies in Applied Bayesian Data Science https://doi.org/10.1007/978-3-030-42553-1_3	book	January 2020
Prediction and uncertainty quantification of SAFARI-1 axial neutron flux profiles with neural networks Moloko, Lesego E.; Bokov, Pavel M.; Wu, Xu Annals of Nuclear Energy, Vol. 188 https://doi.org/10.1016/j.anucene.2023.109813	journal	August 2023
Developing semi-supervised latent dynamic variational autoencoders to enhance prediction performance of product quality Lee, Yi Shan; Chen, Junghui Chemical Engineering Science, Vol. 265 https://doi.org/10.1016/j.ces.2022.118192	journal	January 2023
Functional PCA and deep neural networks-based Bayesian inverse uncertainty quantification with transient experimental data Xie, Ziyu; Yaseen, Mahmoud; Wu, Xu Computer Methods in Applied Mechanics and Engineering, Vol. 420 https://doi.org/10.1016/j.cma.2023.116721	journal	February 2024
Two-step residual-error based approach for anomaly detection in engineering systems using variational autoencoders González-Muñiz, Ana; Díaz, Ignacio; Cuadrado, Abel A. Computers and Electrical Engineering, Vol. 101 https://doi.org/10.1016/j.compeleceng.2022.108065	journal	July 2022
Application of generative deep learning to predict temperature, flow and species distributions using simulation data of a methane combustor Laubscher, Ryno; Rousseau, Pieter International Journal of Heat and Mass Transfer, Vol. 163 https://doi.org/10.1016/j.ijheatmasstransfer.2020.120417	journal	December 2020
Imbalanced data fault diagnosis method for nuclear power plants based on convolutional variational autoencoding Wasserstein generative adversarial network and random forest Guo, Jun; Wang, Yulong; Sun, Xiang Nuclear Engineering and Technology, Vol. 56, Issue 12 https://doi.org/10.1016/j.net.2024.07.015	journal	December 2024
Inverse uncertainty quantification using the modular Bayesian approach based on Gaussian Process, Part 2: Application to TRACE Wu, Xu; Kozlowski, Tomasz; Meidani, Hadi Nuclear Engineering and Design, Vol. 335 https://doi.org/10.1016/j.nucengdes.2018.06.003	journal	August 2018
Deep generative modeling-based data augmentation with demonstration using the BFBT benchmark void fraction datasets Alsafadi, Farah; Wu, Xu Nuclear Engineering and Design, Vol. 415 https://doi.org/10.1016/j.nucengdes.2023.112712	journal	December 2023
Flow field prediction of supercritical airfoils via variational autoencoder based deep learning framework Wang, Jing; He, Cheng; Li, Runze Physics of Fluids, Vol. 33, Issue 8 https://doi.org/10.1063/5.0053979	journal	August 2021
Quantification of Deep Neural Network Prediction Uncertainties for VVUQ of Machine Learning Models Yaseen, Mahmoud; Wu, Xu Nuclear Science and Engineering, Vol. 197, Issue 5 https://doi.org/10.1080/00295639.2022.2123203	journal	November 2022
Variational Inference: A Review for Statisticians Blei, David M.; Kucukelbir, Alp; McAuliffe, Jon D. Journal of the American Statistical Association, Vol. 112, Issue 518 https://doi.org/10.1080/01621459.2017.1285773	journal	July 2016
Unsupervised Real Image Super-Resolution via Generative Variational AutoEncoder Liu, Zhi-Song; Siu, Wan-Chi; Wang, Li-Wen 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) https://doi.org/10.1109/CVPRW50498.2020.00229	conference	June 2020
The variational approximation for Bayesian inference Tzikas, Dimitris G.; Likas, Aristidis C.; Galatsanos, Nikolaos P. IEEE Signal Processing Magazine, Vol. 25, Issue 6 https://doi.org/10.1109/MSP.2008.929620	journal	November 2008
Inference-Reconstruction Variational Autoencoder for Light Field Image Reconstruction Han, Kang; Xiang, Wei IEEE Transactions on Image Processing, Vol. 31 https://doi.org/10.1109/TIP.2022.3197976	journal	January 2022
Variational Autoencoders for Data Augmentation in Clinical Studies Papadopoulos, Dimitris; Karalis, Vangelis D. Applied Sciences, Vol. 13, Issue 15 https://doi.org/10.3390/app13158793	journal	July 2023

Similar Records

Predicting critical heat flux with uncertainty quantification and domain generalization using conditional variational autoencoders and deep neural networks

Journal Article · Mon Apr 28 00:00:00 EDT 2025 · Annals of Nuclear Energy · OSTI ID:2563800

Physics-based hybrid machine learning for critical heat flux prediction with uncertainty quantification

Journal Article · Mon Jul 14 00:00:00 EDT 2025 · Applied Thermal Engineering · OSTI ID:2571909

Related Subjects

Bayesian neural network
Conformal prediction
Data augmentation
Uncertainty quantification
Variational autoencoders

An investigation on machine learning predictive accuracy improvement and uncertainty reduction using VAE-based data augmentation

Citation Formats

References (16)

Similar Records

Related Subjects