Improved FBG In Situ Monitoring Method for Epoxy Curing Shrinkage via Macromolecular Percolation and Dynamic Calibration

In situ monitoring of curing shrinkage using fiber Bragg grating (FBG) sensors provides critical parameters for predicting residual stresses. However, as components evolve toward microminiaturization, reduced geometric boundaries trigger severe free‐edge effects and interfacial shear lag, drastically limiting FBG measurement accuracy. To address this limitation, this study proposes a full‐process dynamic calibration framework for curing shrinkage. Combined with the Kamal–Sourour kinetics model, rotational rheology was utilized to resolve the cross‐linking viscosity divergence and determine the gelation point. On this basis, a temperature‐compensated FBG sensing platform deconstructed the dynamic thermodynamic competition mechanism, validated rigorously via multiple thermodynamic analysis methods. Crucially, finite element analysis elucidated the influence of physical boundaries on strain transfer efficiency. By incorporating macromolecular percolation theory, the continuous mapping of the time‐varying elastic modulus was achieved to establish a dynamic calibration model. Evolutionary laws indicate that a reduction in the radial dimension of the epoxy resin causes a significant decrease in strain energy, forcing most curing shrinkage strain to dissipate entirely within a narrow radial shear gradient. Through this dynamic calibration, the curing‐induced residual strain measured in the small‐scale specimen was corrected from 3173.17 × 10 −6 to 4501.57 × 10 −6 , reducing the relative deviation from the large‐scale reference from 34.91% to 7.66%. This work effectively decouples the reliance of FBG in situ monitoring on material geometric boundary scales, offering a universal framework for high‐precision manufacturing optimization.