Abstract
Engineering systems are growing in complexity, requiring increasingly intelligent and flexible methods to account for and predict uncertainties in service. This paper presents a framework for dynamic uncertainty prediction under limited data (UPLD). Spatial geometry is incorporated with LSTM networks to enable real-time multistep prediction of quantitative and qualitative uncertainty over time. Validation is achieved through two case studies. Results demonstrate robust prediction of trends in limited and dynamic uncertainty data with parallel determination of geometric symmetry at each time unit. Future work is recommended to explore alternative network architectures suited to limited data scenarios.
Highlights
The growing complexity in engineering systems manifests a range of uncertainty surrounding in-service maintenance
This paper presented a framework to predict dynamic un certainty exhibited under limited data (UPLD) for the maintenance of increasingly complex engineering systems
These uncertainties arise as a result of data quality and availability, operating conditions and assumptions made surrounding maintenance
Summary
The growing complexity in engineering systems manifests a range of uncertainty surrounding in-service maintenance. The steps were developed from emerging studies in literature utilising LSTM networks to forecast time-series data, extended to consider the geometric sym metry between input variances to improve prediction robustness under limited data. Other factor holds constant up to year 23 before an unexpected dive, which the network was not able to account for in the prediction These sudden changes and the scale in the observed variance data directly impact the mean prediction error, causing the high variation in MAPE and RMSE over the test period. As for Section “Case study 1: US SAR data”, predictions are robust as they reflect observed trends despite outliers and limited data on which to train [57]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: CIRP Journal of Manufacturing Science and Technology
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
