A Fiber Bragg Grating (FBG) biosensor with good biocompatibility promises label-free, anti-electromagnetic interference, and non-destructive detection of stress monitoring, it holds the potential to rapidly identify the stress changes in microfluidics systems. Based on signal processing, the correlation between wavelength shift in FBG and flowrate is discussed, an alternative to manual processing of wavelength offset calculation is analyzed, and the automatic selection of FBG wavelength change under high noise conditions is realized. In addition, we propose some technologies of intelligent processing algorithms such as machine learning (BP-back propagation neural network) and moving average. Finally, the feasibility and stability of the signal processing method are verified by experimental validation and multiple data sets testing. This study provides a promising intelligent information processing strategy for rapid and accurate stress sensing of microfluidics systems.