Journal
IEEE-ASME TRANSACTIONS ON MECHATRONICS
Volume 27, Issue 6, Pages 4686-4696Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMECH.2022.3164022
Keywords
Deep learning models; fault diagnosis; interdigital electrodes; prebent membrane (PM); rotating machinery; triboelectric nanogenerator (TENG)
Categories
Funding
- National Science Foundation of China [11872222]
- State Key Laboratory of Tribology [SKLT2021D11]
Ask authors/readers for more resources
A prebent membrane-based triboelectric nanogenerator (PM-TENG) was proposed for harvesting rotational energy and applied to fault diagnosis in rotating machinery. It was demonstrated that the PM-TENG has self-powered characteristics and can effectively drive micropowered electronic devices. By identifying fault characteristic frequencies using the PM-TENG's output current model and applying signal processing and machine learning algorithms, fault classification of rotating machinery can be achieved with high accuracy.
A prebent membrane-based triboelectric nanogenerator (PM-TENG) was proposed to harvest rotational energy; its application in diagnosing faults in rotating machinery was then explored. A PM array was realized by pressing two sides of each arch membrane into the radial grooves of a rotor disk, and a PM-TENG was formed together with a stator disk with interdigital electrodes pasted on the surface. Variations in output voltage and current with load resistance were tested, and the effects of design parameters on the output characteristics were discussed. It is demonstrated that the proposed PM-TENG has self-powered characteristics by charging the load capacitor and effectively driving micropowered electronic devices. Two test rigs were constructed for fault diagnosis tests in the rotating machinery. Fault characteristic frequencies were identified using the output current model of the PM-TENG. Fast Fourier transform and deep learning models were used to classify only bearing and gear-bearing hybrid faults, respectively. The results showed that the PM-TENG output current can be used to diagnose typical faults in rotating machinery; the classification accuracy exceeded 92%, which is only slightly lower than that based on the vibration signal. The proposed PM-TENG has application potential for rotating machinery fault diagnosis.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available