Due to the importance of sensors in control strategy and safety, early detection of faults in sensors has become a key point to improve the availability of railway traction drives. The presented sensor fault reconstruction is based on sliding mode observers and equivalent injection signals, and it allows detecting defective sensors and isolating faults. Moreover, the severity of faults is provided. The proposed on-board fault reconstruction has been validated in a hardware-in-the-loop platform, composed of a real-time simulator and a commercial traction control unit for a tram. Low computational resources, robustness to measurement noise, and easiness to tune are the main requirements for industrial acceptance. As railway applications are not safety-critical systems, compared to aerospace applications, a fault evaluation procedure is proposed, since there is enough time to perform diagnostic tasks. This procedure analyses the fault reconstruction in the steady state, delaying the decision-making in some seconds, but minimising false detections.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6069394 | PMC |
| http://dx.doi.org/10.3390/s18071998 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
High-resolution investigations of fault architecture in space and time.
Sci Rep
January 2025
Dipartimento di Scienze Biologiche, Geologiche ed Ambientali, Università di Bologna, Bologna, Italy.
Heterogeneous fault architecture affects crustal seismotectonics and fluid migration. When studying it, we commonly rely on static conceptual models that generally overlook the absolute time dimension of fault (re)activation. Heterogenous faults, however, represent the end-result of protracted, cumulative and intricate deformation histories.
View Article and Find Full Text PDFSystem identification and fault reconstruction in solar plants via extended Kalman filter-based training of recurrent neural networks.
ISA Trans
January 2025
Dept. de Ingeniería de Sistemas y Automática, University of Seville, Camino de los Descubrimientos, no number E-41092, Seville, Spain. Electronic address:
This article proposes using the extended Kalman filter (EKF) for recurrent neural network (RNN) training and fault estimation within a parabolic-trough solar plant. The initial step involves employing an RNN to model the system. Given the challenge of fault discernibility in the collectors, parallel EKFs are employed to reconstruct the parameters of the faults.
View Article and Find Full Text PDFNovel GSIP: GAN-based sperm-inspired pixel imputation for robust energy image reconstruction.
Sci Rep
January 2025
Department of Computer Science, Faculty of Computers and Informatics, Kafrelsheikh University, Kafrelsheikh, Egypt.
Missing pixel imputation is a critical task in image processing, where the presence of high percentages of missing pixels can significantly degrade the performance of downstream tasks such as image segmentation and object detection. This paper introduces a novel approach for missing pixel imputation based on Generative Adversarial Networks (GANs). We propose a new GAN architecture incorporating an identity module and a sperm motility-inspired heuristic during filtration to optimize the selection of pixels used in reconstructing missing data.
View Article and Find Full Text PDFSeq2Seq-based GRU autoencoder for anomaly detection and failure identification in coal mining hydraulic support systems.
Sci Rep
January 2025
School of Mines, China University of Mining and Technology, Xuzhou, China.
In coal mining operations, the stable operation of hydraulic supports is crucial for ensuring mine safety. However, the nonlinear, non-stationary characteristics and noise interference in hydraulic support pressure data pose significant challenges for anomaly detection and fault diagnosis. This study proposes an anomaly detection and failure identification method based on Gated Recurrent Unit Autoencoder (GRU-AE), aimed at achieving anomaly detection in hydraulic support pressure data and equipment failure early warning.
View Article and Find Full Text PDFThrust-dominated unilateral rupture of a blind listric fault associated with the 2024 Hualien earthquake.
Sci Rep
December 2024
Department of Earth and Space Sciences, Southern University of Science and Technology, Shenzhen, China.
The 2024 Hualien M 7.4 earthquake struck the Longitudinal Valley, which accommodates the partial collision between the Eurasian and Philippine Sea plates. As the most significant event in Taiwan since the 1999 Chi-Chi M 7.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!