Abstract
A unique multi-axis vibration sensor system based on a flexible Fiber Bragg Grating (FBG) hinge is presented in this paper. Derivation of sensitivity formulae for longitudinal and transverse vibration offers insights into the sensor’s behavior in response to stimuli that are directed in multiple directions. The sensor design is improved via a multi-objective optimization technique, which minimizes interference and increases sensitivity. By tackling cross-talk and interference, a Random Forest-based recognition model substantially increases the accuracy of vibration signal identification. High sensitivities are seen in the 100–1000 Hz frequency range: 8.847 pm/g, 11.71 pm/g, and 7.687 pm/g. Impressive lateral anti-interference identification rates of 99.58% and 98.75% are attained by our Random Forest model, proving its resilience in a variety of monitoring circumstances. These results highlight the improved dependability and performance of FBG hinge-type accelerometers.
Funding Information
  • Research Program of Sanya Yazhou Bay Science and Technology City (SKJC-2020-01-016)
  • Key Project of the Ministry of Industry and Information Technology of the People’s Republic of China (TC220A04W-5, 187)
  • Fundamental Research Funds for the Central Universities (WUT-104972024RSCbs0048)

This publication has 26 references indexed in Scilit: