Triboelectric Nanogenerator Assisted by Machine Learning

Dongfang Yang,Yuchao Shang,Zhaoting Li,Kai Wang

DOI: https://doi.org/10.1021/acsaelm.3c01043

IF: 4.494

2023-11-21

ACS Applied Electronic Materials

Abstract:In the era of the Internet of Things (IoT), the development of an intelligent sensor system with a sustainable power supply, convenient deployment, and versatile use has emerged as an urgent challenge. Triboelectric nanogenerators (TENGs) stand out as some of the most promising energy solutions in this context. They possess the capability of directly transforming mechanical stimuli into electrical signals, making them highly valuable for self-powered sensor applications. Sensors based on a TENG exhibit the advantages of uncomplicated structures and remarkable instantaneous power density, thereby serving as a critical component in the construction of intelligent sensor systems. In parallel, machine learning (ML), with its attributes of cost-effectiveness, short development cycles, and robust data processing and predictive capabilities, has matured over more than five decades to meet the research requirements of various disciplines. ML demonstrates exceptional prowess in managing the substantial volume of electrical signals generated by a TENG. This paper presents a comprehensive overview of recent advancements in signal processing and intelligent identification within sensor systems based on a TENG. The technical characteristics and research status of this convergence are explored across five key domains: traffic safety, environmental monitoring, information security, motion tracking, and human–computer interaction. Finally, this paper delves into the current challenges and future development trends within this field, offering an analytical perspective on how to enhance its prospects and ultimately opening up a broader landscape for potential applications. We believe that the fusion of ML technology and TENG sensors will catalyze the rapid evolution of intelligent sensor networks in the foreseeable future, providing valuable insights for other research domains as well.

materials science, multidisciplinary,engineering, electrical & electronic