High‐Performance Transparent, Deformable, and Recoverable Biomimetic Stevia–PVA Hydrogel Triboelectric Nanogenerator with Machine Learning‐Assisted Motion Recognition
Advanced Materials · 2026-07 · 38 (41)
DOI: 10.1002/adma.73030 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advanced Materials, online erschienen am 08.04.2026. Verfasst von Thien Trung Luu und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanical Engineering, College of Engineering, Sungkyunkwan University, Suwon, Gyeonggi, South Korea. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Research Foundation of Korea, Korea government, National Research Foundation.
Abstract (Originalsprache)
Rapid breakthroughs in IoT and AI have raised demand for portable, self‐powered, flexible sensor devices. Hydrogels with high conductivity, mechanical tunability, environmental adaptability, and biocompatibility are a clever way to develop flexible sensors for triboelectric nanogenerators. TENG application is limited by the paucity of suitable biomaterials and the need for highly conductive fillers such 2D materials, which trade off transparency, output, and sensing. A unique, very transparent, highly stretchable, high‐output performance biomimetic stevia/PVA hydrogel‐based triboelectric nanogenerator (S‐TENG) is investigated to overcome this issue. Due to its abundant dynamic hydrogen bonding, cost‐effective biomimetic stevia is added to polyvinyl alcohol (PVA) to increase hydrogel cross‐linking and crystalline domains. These structural advancements give the S‐hydrogel 2–5 times the mechanical strength and 3–8 times the electrical output of 2D‐, bio‐, and transparent‐material‐based TENGs, while maintaining transparency. The S‐hydrogel may be recycled and recovered by water‐assisted dissolution and re‐gelation, keeping its voltage output. The improved S‐TENG is a self‐powered sensor for various human motions with great sensitivity and a 13‐ms reaction time. The XGBoost method had the greatest classification accuracy of 95.29% among eleven machine learning models, showing the promise of self‐powered sensors for many applications.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Mechanical Engineering, College of Engineering, Sungkyunkwan University, Suwon, Gyeonggi, South Korea.
- Department of Chemical Engineering (Integrated Engineering), Kyung Hee University, Yongin-si, Gyeonggi-do, Republic of Korea.
- Department of Future Energy Engineering, Sungkyunkwan University, Suwon, Republic of Korea.
ORCID-Profile: Kyungwho Choi, Dukhyun Choi
Förderung
- National Research Foundation of Korea
- Korea government
- National Research Foundation
Zitieren
APA
Luu, T. T., Quang, B. M., Pham, T. M., Kim, J., Choi, K., & Choi, D. (2026). High‐Performance Transparent, Deformable, and Recoverable Biomimetic Stevia–PVA Hydrogel Triboelectric Nanogenerator with Machine Learning‐Assisted Motion Recognition. Advanced Materials, 38(41). https://doi.org/10.1002/adma.73030
Vancouver
Luu TT, Quang BM, Pham TM, Kim J, Choi K, Choi D. High‐Performance Transparent, Deformable, and Recoverable Biomimetic Stevia–PVA Hydrogel Triboelectric Nanogenerator with Machine Learning‐Assisted Motion Recognition. Advanced Materials. 2026;38(41). doi:10.1002/adma.73030
BibTeX
@article{Luu2026,
author = {Luu, Thien Trung and Quang, Bui Minh and Pham, Toan Minh and Kim, Jinsoo and Choi, Kyungwho and Choi, Dukhyun},
title = {High‐Performance Transparent, Deformable, and Recoverable Biomimetic Stevia–PVA Hydrogel Triboelectric Nanogenerator with Machine Learning‐Assisted Motion Recognition},
journal = {Advanced Materials},
year = {2026},
volume = {38},
number = {41},
doi = {10.1002/adma.73030}
}
EUVIMED