Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic Systems

Uncertainty quantification (UQ) is critical for modeling complex dynamic systems, ensuring robustness and interpretability. This study extends Physics-Guided Bayesian Neural Networks (PG-BNNs) to enhance model robustness by integrating physical laws into Bayesian frameworks. Unlike Artificial Neural...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinyue Xu, Julian Wang
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Forecasting
Subjects:
Online Access:https://www.mdpi.com/2571-9394/7/1/9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850204460443238400
author Xinyue Xu
Julian Wang
author_facet Xinyue Xu
Julian Wang
author_sort Xinyue Xu
collection DOAJ
description Uncertainty quantification (UQ) is critical for modeling complex dynamic systems, ensuring robustness and interpretability. This study extends Physics-Guided Bayesian Neural Networks (PG-BNNs) to enhance model robustness by integrating physical laws into Bayesian frameworks. Unlike Artificial Neural Networks (ANNs), which provide deterministic predictions, and Bayesian Neural Networks (BNNs), which handle uncertainty probabilistically but struggle with generalization under sparse and noisy data, PG-BNNs incorporate the laws of physics, such as governing equations and boundary conditions, to enforce physical consistency. This physics-guided approach improves generalization across different noise levels while reducing data dependency. The effectiveness of PG-BNNs is validated through a one-degree-of-freedom vibration system with multiple noise levels, serving as a representative case study to compare the performance of Monte Carlo (MC) dropout ANNs, BNNs, and PG-BNNs across interpolation and extrapolation domains. Model accuracy is assessed using Mean Squared Error (MSE), Mean Absolute Percentage Error (MAE), and Coefficient of Variation of Root Mean Square Error (CVRMSE), while UQ is evaluated through 95% Credible Intervals (CIs), Mean Prediction Interval Width (MPIW), the Quality of Confidence Intervals (QCI), and Coverage Width-based Criterion (CWC). Results demonstrate that PG-BNNs can achieve high accuracy and good adherence to physical laws simultaneously, compared to MC dropout ANNs and BNNs, which confirms the potential of PG-BNNs in engineering applications related to dynamic systems.
format Article
id doaj-art-c77b9b5b9b6f4883aa91594a086a2de3
institution OA Journals
issn 2571-9394
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Forecasting
spelling doaj-art-c77b9b5b9b6f4883aa91594a086a2de32025-08-20T02:11:17ZengMDPI AGForecasting2571-93942025-02-0171910.3390/forecast7010009Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic SystemsXinyue Xu0Julian Wang1Department of Architectural Engineering, Pennsylvania State University, State College, PA 16802, USADepartment of Architectural Engineering, Pennsylvania State University, State College, PA 16802, USAUncertainty quantification (UQ) is critical for modeling complex dynamic systems, ensuring robustness and interpretability. This study extends Physics-Guided Bayesian Neural Networks (PG-BNNs) to enhance model robustness by integrating physical laws into Bayesian frameworks. Unlike Artificial Neural Networks (ANNs), which provide deterministic predictions, and Bayesian Neural Networks (BNNs), which handle uncertainty probabilistically but struggle with generalization under sparse and noisy data, PG-BNNs incorporate the laws of physics, such as governing equations and boundary conditions, to enforce physical consistency. This physics-guided approach improves generalization across different noise levels while reducing data dependency. The effectiveness of PG-BNNs is validated through a one-degree-of-freedom vibration system with multiple noise levels, serving as a representative case study to compare the performance of Monte Carlo (MC) dropout ANNs, BNNs, and PG-BNNs across interpolation and extrapolation domains. Model accuracy is assessed using Mean Squared Error (MSE), Mean Absolute Percentage Error (MAE), and Coefficient of Variation of Root Mean Square Error (CVRMSE), while UQ is evaluated through 95% Credible Intervals (CIs), Mean Prediction Interval Width (MPIW), the Quality of Confidence Intervals (QCI), and Coverage Width-based Criterion (CWC). Results demonstrate that PG-BNNs can achieve high accuracy and good adherence to physical laws simultaneously, compared to MC dropout ANNs and BNNs, which confirms the potential of PG-BNNs in engineering applications related to dynamic systems.https://www.mdpi.com/2571-9394/7/1/9uncertainty quantificationphysics-guided neural networkspredictive capabilityBayesian neural networkvibration dynamics
spellingShingle Xinyue Xu
Julian Wang
Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic Systems
Forecasting
uncertainty quantification
physics-guided neural networks
predictive capability
Bayesian neural network
vibration dynamics
title Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic Systems
title_full Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic Systems
title_fullStr Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic Systems
title_full_unstemmed Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic Systems
title_short Comparative Analysis of Physics-Guided Bayesian Neural Networks for Uncertainty Quantification in Dynamic Systems
title_sort comparative analysis of physics guided bayesian neural networks for uncertainty quantification in dynamic systems
topic uncertainty quantification
physics-guided neural networks
predictive capability
Bayesian neural network
vibration dynamics
url https://www.mdpi.com/2571-9394/7/1/9
work_keys_str_mv AT xinyuexu comparativeanalysisofphysicsguidedbayesianneuralnetworksforuncertaintyquantificationindynamicsystems
AT julianwang comparativeanalysisofphysicsguidedbayesianneuralnetworksforuncertaintyquantificationindynamicsystems