Vibrational Triboelectric Nanogenerator-Based Multinode Self-Powered Sensor Network for Machine Fault Detection
Wenjian Li, Yaoyao Liu, Shuwei Wang, Wei Li, Guoxu Liu, Junqing Zhao, Xiaohan Zhang, Chi Zhang
Abstract
Physical parameter sensing largely benefits the lifetime and operational costs of machines and has been widely used for machine fault detection. Herein, in this article, we developed a multinode sensor network, which is fully self-powered by harvesting mechanical vibration energy, to establish a machine fault detection system. A multilayered vibrational triboelectric nanogenerator (V-TENG) was designed to scavenge energy from working machines. Triggered by a vibration motion with the frequency of 8 Hz, the V-TENG can generate an output with power density of 3.33 mW/m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> . With a power management module, the microcontrol unit integrated with sensors and a wireless transmitter can be continuously powered by the V-TENG to construct a self-powered vibration sensor node (SVSN). A supporting vector machine algorithm-based machine fault detection system was then established through a three-SVSN network by acquiring acceleration and temperature data from the working machine. Based on the system, different working conditions of the machine were recognized with an accuracy of 83.6%. The TENG-based SVSN for machine fault detection has demonstrated wide prospects in production monitoring, intelligent manufacturing, and smart factory. Moreover, the proposed self-powered sensor network has great potential and wide application in the era of distributed Internet of Things, artificial intelligence, and big data.