A High-Precision 3-D Vibration Acceleration Sensor Based on FBGs

In this article, a 3-D vibration acceleration sensor based on fiber Bragg grating (FBG) is proposed. The sensor, with only four FBGs, adopts a spatially layered sensing structure and achieves vibration acceleration measurements as well as a temperature compensation function. This document delves into the principles of FBG sensing and mechanics to understand the relationship between the acceleration this sensor measures and the wavelength drift. Finite element analysis (FEA) was used to optimize the pasting position of the FBG and to understand the resonance frequency of the sensor. The experimental results show that the operating frequency ranges of the sensors for the X-, Y-, and Z-axes are 0–600, 0–700, and 0–600 Hz, respectively, with sensitivities are 7.85, 8.15, and 6.65 pm/g, respectively. With repeatability errors of 2.32%, 1.54%, and 1.06% and a stability test error of 1.124%, the sensor proved to have high stability. The neural network model was constructed using an extreme learning machine (ELM) based on the whale optimization algorithm (WOA). The model is optimized using WOA to obtain its optimal initial weights and thresholds. After the algorithm’s optimization, the maximum linear error of the sensor’s three axes is 1.72%, and the maximum coupling error is 0.78%, proving that the sensor has high measurement accuracy. Combined with the sensor’s stability and measurement accuracy analysis, it proves that the sensor can not only measure 3-D vibration acceleration but also has high accuracy.