Deep Learning-Based Remaining Useful Life Prediction for Mechatronical Components
Typical mechanical and electronic (mechatronical) components, including gearboxes, bearings, motors, and principal axes, are used extensively in a variety of engineering fields. Due to their harsh operating conditions, mechatronical components are often subject to faults, such as cracking, wear, pitting, spalling, fracturing, looseness, friction between rotor and stator, broken bars, and short circuiting, etc., resulting in huge economic losses and serious injuries or even casualties. Consequently, it is of great concern to intelligently predict the remaining useful life (RUL) of key mechatronical components to ensure that the equipment's reliable operation and making optimal maintenance decisions.
With the rapid development of industrial big data and artificial intelligence technology, deep learning-based fault prognosis and diagnosis methods have attracted a lot of attention, as they can greatly reduce the interference of expertise and enhance diagnostic efficiency. When there is a significant amount of annotated training data, deep learning (DL) models demonstrate appealing performance. However, high-quality labeled life-cycle data are still rare in actual engineering practice, and various monitoring data from different but similar components and the varied working conditions have prominent distribution discrepancy; thus, DL-based RUL prediction methods still face great challenges. Aiming to address these issues, transfer learning has been proposed to improve the accuracy and generalization performance of RUL prediction, becoming a research hotspot.
In order to facilitate the development of intelligent RUL prediction, we are organizing a Topic titled “Deep Learning-Based Remaining Useful Life Prediction for Mechatronical Components”. The Topic will be published in the journals Machines, Applied Sciences, Sensors, Electronics and Chips. This topic hopes to attract studies including health indicator construction, remaining useful life (RUL) prediction for mechanical and electronic components, transfer RUL prediction, interpretable deep learning RUL prediction models, federal learning-based RUL prediction, etc.
Topics mainly include:
- Supervised health indicator construction;
- Unsupervised health indicator construction;
- Time series prediction models based on recurrent neural works;
- RUL prediction based on pattern recognition;
- Transfer RUL prediction based on domain adaptation;
- Interpretable deep learning models for RUL prediction;
- Federal learning-based RUL prediction.
Prof. Dr. Jun Wu
Dr. Zhaojun Steven Li
|Journal Name||Impact Factor||CiteScore||Launched Year||First Decision (median)||APC|
|2.7||4.5||2011||15.8 Days||CHF 2300||Submit|
|-||-||2022||15.0 days *||CHF 1000||Submit|
|2.9||4.7||2012||15.8 Days||CHF 2200||Submit|
|2.6||2.1||2013||15 Days||CHF 2400||Submit|
|3.9||6.8||2001||16.4 Days||CHF 2600||Submit|
* Median value for all MDPI journals in the first half of 2023.
Preprints is a platform dedicated to making early versions of research outputs permanently available and citable. MDPI journals allow posting on preprint servers such as Preprints.org prior to publication. For more details about reprints, please visit https://www.preprints.org.