Coaxial Flexible Fiber‐Shaped Triboelectric Nanogenerator Assisted by Deep Learning for Self‐Powered Vibration Monitoring

• Published in: Small (Weinheim an der Bergstrasse, Germany) Vol. 20; no. 15; pp. e2307680 - n/a
• Main Authors: Zhao, Cong, Du, Taili, Ge, Bin, Xi, Ziyue, Qian, Zian, Wang, Yawei, Wang, Junpeng, Dong, Fangyang, Shen, Dianlong, Zhan, Zhenhao, Xu, Minyi
• Format: Journal Article
• Language: English
• Published: Germany Wiley Subscription Services, Inc 01.04.2024
• Subjects: Core-shell structure; Deep learning; Diesel engines; Error detection; fiber‐shaped sensors; Industrial applications; Industry 4.0; Nanogenerators; self‐powered sensors; triboelectric nanogenerators; Vibration monitoring; vibration sensors; triboelectric nanogenerators; self‐powered sensors; fiber‐shaped sensors; vibration sensors
• DOI: 10.1002/smll.202307680
• ISSN: 1613-6810; 1613-6829

Self‐powered vibration sensor is highly desired for distributed and continuous monitoring requirements of Industry 4.0. Herein, a flexible fiber‐shaped triboelectric nanogenerator (F‐TENG) with a coaxial core‐shell structure is proposed for the vibration monitoring. The F‐TENG exhibits higher adaptability to the complex surfaces, which has an outstanding application prospect due to vital compensation for the existing rigid sensors. Initially, the contact characteristics between the dielectric layers, that related to the perceiving performance of the TENG, are theoretically analyzed. Such a TENG with 1D structure endows high sensitivity, allowing for accurately responding to a wide range of vibration frequencies (0.1 to 100 Hz). Even applying to the real diesel engine, the error in detecting the vibration frequencies is only 0.32% compared with the commercial vibration sensor, highlighting its potential in practical application. Further, assisted by deep learning, the recognition accuracy in monitoring nine operating conditions of the system achieves 97.87%. Overall, the newly designed F‐TENG with the merits of high‐adaptability, cost‐efficiency, and self‐powered, has offered a promising solution to fulfill an extensive range of vibration sensing applications in the future. A coaxial flexible fiber‐shaped triboelectric nanogenerator (F‐TENG) is developed for self‐powered vibration sensor. The mechanical and electrical characteristics of the F‐TENG are theoretically analyzed. And the optimized device can detect vibration in broadband frequency range. Assisted by deep‐learning, the proposed TENG shows promising potential in monitoring operational conditions and identifying faults of the system.