Using Genetic Algorithm for Parameter Reconstruction of a Chirped Fiber Bragg Grating

We propose a novel method for multiple parameters and phase reconstruction of a chirped Fiber Bragg Grating (FBG) through two thermally-modulated reflection intensity spectra and genetic algorithm. Two temperature distributions are used to modulate the chirped FBG’s intensity spectrum in order to synthesize FBG’s parameters through a genetic algorithm. Thus, the phase response can be reconstructed from the complete FBG’s parameters. The advantages of the proposed method are simple, low cost, and nondestructive characterization. This method can also resolve the ambiguity of chirped FBG’s with opposite structures. As a result, the experimental results show that multiple parameters and the phase response of a chirped FBG can be reconstructed correctly. However, the drawbacks of this approach are long computation time about 1 Hour in 2.4 GHz PC for acceptable results and the accuracy of temperature distribution measurement along the substrate is needed. After finding these physical parameters, the arbitrary distributed strain sensing application is easy to accomplish by using two fiber Bragg gratings sensing techniques. We also proposes and verifies a simple, convenient, and low cost method to inversely measure arbitrary strain distributions by applying a genetic algorithm approach to analyze the reflection intensity spectra of two fiber Bragg gratings (FBGs). The proposed method involves bonding one uniform FBG and one chirped FBG to the same location of the structure of interest such that they both encounter the same strain field. The arbitrary strain distribution within the fiber gratings is then determined inversely from the two Bragg spectra by means of a genetic algorithm population-based optimization process.