Mechanical-computing metastructure for self-powered vibration sensing
Hongbin Pan, Jiaxi Zhou, K. Wang, Qiang Wang, Dongguo Tan
Abstract
Triboelectric nanogenerator-based vibration sensors (TVSs) present a promising approach for online vibration monitoring , offering early detection of excessive vibrations that pose risks to the functionality of devices and infrastructure. However, the structural design of most current TVSs lacks consideration for information processing capabilities, resulting in limited detection bandwidth and displacement signal fidelity. This study first introduces a quasi-zero-stiffness (QZS) mechanical-computing metastructure aimed at enhancing the structural intelligence of TVSs and extending their vibration signal detection bandwidth. The optimization of the QZS metastructure involved adjusting the length ratio and inclination angle of the zero-order polynomial beam to achieve ultra-low resonant frequencies. The results indicate that the TVS based on the mechanical-computing metastructure (MCM-TVS) is capable of efficiently detecting complex random signals, achieving an average measurement accuracy above 85 % for the spectral components of random vibration signals. Furthermore, the sensor exhibited remarkable linearity (99 %) in measuring vibration amplitude and maintained consistent sensitivity (with less than 25.8 % fluctuation), facilitating the lossless conversion of displacement to electrical signals. Consequently, this novel approach of incorporating an MCM into TVS design offers substantial insights for the advancement of high-performance vibration signal monitoring.