Suppression of sweeping fluctuation of Fabry-Perot filter in fiber Bragg grating interrogation using PSO-based self-adaptive sampling

Abstract

Random fluctuations of the nonlinearity curves of Fabry–Perot (F–P) tunable filters cause instabilities in fiber Bragg grating (FBG) interrogation systems. FBG sensors are widely used in structural health monitoring (SHM), and suppressing random fluctuations can provide engineers with more precise information on the conditions of structures. While F–P etalon and acetylene (C2H2) are most commonly used to calibrate the wavelength of the F-P filter, in medium–high sweeping situations, they suffer from drastic fluctuations in the interrogation results. In addition, these two methods require more interrogation components and render the interrogation system more complex and expensive. This study investigates the fluctuation of interrogation errors caused by sampling frequency. A novel self-adaptive sampling approach based on particle swarm optimization (PSO) is proposed to automatically search the optimal sampling frequency with the smallest fluctuation. The experimental results show that the proposed approach successfully detected the optimal sampling frequency. The drastic fluctuations of micro strains caused by high sampling frequency are avoided completely, and the stability of interrogation system is guaranteed. The proposed approach is thus helpful for suppressing sweeping fluctuations by appropriately choosing the optimal sampling frequency.