skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


This content will become publicly available on March 1, 2026

Title: Harnessing Bayesian Deep Learning to Tackle Unseen and Uncertain Scenarios in Diagnosis of Machinery Systems
Abstract Direct inverse analysis of faults in machinery systems such as gears using first principle is intrinsically difficult, owing to the multiple time- and length-scales involved in vibration modeling. As such, data-driven approaches have been the mainstream, whereas supervised trainings are deemed effective. Nevertheless, existing techniques often fall short in their ability to generalize from discrete data labels to the continuous spectrum of possible faults, which is further compounded by various uncertainties. This research proposes an interpretability-enhanced deep learning framework that incorporates Bayesian principles, effectively transforming convolutional neural networks (CNNs) into dynamic predictive models and significantly amplifying their generalizability with more accessible insights of the model's reasoning processes. Our approach is distinguished by a novel implementation of Bayesian inference, enabling the navigation of the probabilistic nuances of gear fault severities. By integrating variational inference into the deep learning architecture, we present a methodology that excels in leveraging limited data labels to reveal insights into both observed and unobserved fault conditions. This approach improves the model's capacity for uncertainty estimation and probabilistic generalization. Experimental validation on a lab-scale gear setup demonstrated the framework's superior performance, achieving nearly 100% accuracy in classifying known fault conditions, even in the presence of significant noise, and maintaining 96.15% accuracy when dealing with unseen fault severities. These results underscore the method's capability in discovering implicit relations between known and unseen faults, facilitating extended fault diagnosis, and effectively managing large degrees of measurement uncertainties.  more » « less
Award ID(s):
2138522
PAR ID:
10584072
Author(s) / Creator(s):
; ;
Publisher / Repository:
The American Society of Mechanical Engineers
Date Published:
Journal Name:
ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering
Volume:
11
Issue:
1
ISSN:
2332-9017
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. (, Proceedings of the ASME 2020 Internal International Symposium on Flexible Automation)
    null (Ed.)
    Condition assessment of machinery components such as gears is important to maintain their normal operations and thus can bring benefit to their life circle management. Data-driven approaches haven been a promising way for such gear condition monitoring and fault diagnosis. In practical situation, gears generally have a variety of fault types, some of which exhibit continuous severities of fault. Vibration data collected oftentimes are limited to reflect all possible fault types. Therefore, there is practical need to utilize the data with a few discrete fault severities in training and then infer fault severities for the general scenario. To achieve this, we develop a fuzzy neural network (FNN) model to classify the continuous severities of gear faults based on the experimental measurement. Principal component analysis (PCA) is integrated with the FNN model to capture the main features of the time-series vibration signals with dimensional reduction for the sake of computational efficiency. Systematic case studies are carried out to validate the effectiveness of proposed methodology. 
    more » « less
  2. ; ; (, International High Performance Buildings Conference)
    Faults in components (valves, sensors, etc.) of radiant floor heating and cooling systems affect the efficiency, cooling and heating capacity as well as the reliability of the system. While various fault detection and diagnostic (FDD) methods have been developed and tested for building systems, FDD algorithms for radiant heating and cooling systems have previously not been available. This paper presents an evolving learning-based FDD approach for a radiant floor heating and cooling system based on growing Gaussian mixture regression (GGMR). The experimental space was controlled with a building automation system (BAS) in which the operating conditions can be monitored, and control parameters can be overridden to desired values. Trend data for normal operation and faulty operation were collected. A total of six fault types with different severities in a radiant floor system were emulated through overriding control parameters. An FDD model based on the GGMR approach was developed with training data and the performance of the model was tested for "known" faults that were including in the training and new "unknown" faults that were implemented in the fault testing. The prediction accuracy for each known fault was extremely high with the lowest prediction accuracy of 98% for one of the faults. The algorithm was successful in detecting the new fault as an unknown state before evolving the model and in diagnosing it as a new fault after evolving the model. 
    more » « less
  3. ; ; ; ; (, 2019 IEEE International Conference on Prognostics and Health Management (ICPHM))
    Early detection of incipient faults is of vital im- portance to reducing maintenance costs, saving energy, and enhancing occupant comfort in buildings. Popular supervised learning models such as deep neural networks are considered promising due to their ability to directly learn from labeled fault data; however, it is known that the performance of supervised learning approaches highly relies on the availability and quality of labeled training data. In Fault Detection and Diagnosis (FDD) applications, the lack of labeled incipient fault data has posed a major challenge to applying these supervised learning techniques to commercial buildings. To overcome this challenge, this paper proposes using Monte Carlo dropout (MC-dropout) to enhance the supervised learning pipeline, so that the resulting neural network is able to detect and diagnose unseen incipient fault examples. We also examine the proposed MC-dropout method on the RP-1043 dataset to demonstrate its effectiveness in indicating the most likely incipient fault types. 
    more » « less
  4. ; ; ; ; ; (, 2019 IEEE International Conference on Prognostics and Health Management (ICPHM))
    Early detection of incipient faults is of vital im- portance to reducing maintenance costs, saving energy, and enhancing occupant comfort in buildings. Popular supervised learning models such as deep neural networks are considered promising due to their ability to directly learn from labeled fault data; however, it is known that the performance of supervised learning approaches highly relies on the availability and quality of labeled training data. In Fault Detection and Diagnosis (FDD) applications, the lack of labeled incipient fault data has posed a major challenge to applying these supervised learning techniques to commercial buildings. To overcome this challenge, this paper proposes using Monte Carlo dropout (MC-dropout) to enhance the supervised learning pipeline, so that the resulting neural network is able to detect and diagnose unseen incipient fault examples. We also examine the proposed MC-dropout method on the RP-1043 dataset to demonstrate its effectiveness in indicating the most likely incipient fault types. 
    more » « less
  5. ; ; ; ; ; ; (, Proceedings of Machine Learning Research)
    In many supervised learning settings, elicited labels comprise pairwise comparisons or rankings of samples. We propose a Bayesian inference model for ranking datasets, allowing us to take a probabilistic approach to ranking inference. Our probabilistic assumptions are motivated by, and consistent with, the so-called Plackett-Luce model. We propose a variational inference method to extract a closed-form Gaussian posterior distribution. We show experimentally that the resulting posterior yields more reliable ranking predictions compared to predictions via point estimates. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email