Strain profile reconstruction of fiber Bragg grating with gradient using chaos genetic algorithm and modified transfer matrix formulation

In this paper, a new heuristic approach based on chaos genetic algorithm and modified transfer matrix method is presented to demodulate the fiber Bragg grating strain sensors. To facilitate accurate calculation of the grating reflected intensity spectrum, the modified transfer matrix approach is applied. The motivation for using the spline smooth function scheme is to provide nice approximation of strain profiles from a scattered data set and overcome the difficulty of calculating the strain gradients in local period function of the modified T-matrix formulation for the piecewise constant strain field assumption. The chaos genetic algorithm is developed to improve the performance of GA and optimize the data of control nodes of the spline smooth function, and more valuably, the reconstructed strain profile is continuous with discretionary spatial resolution. The proposed method is verified through numerical example reconstructions of Bragg grating sensor simulated strain profile cases. (C) 2008 Elsevier B.V. All rights reserved.