Research On FBG Sensing Performance Improving Methods Based On Intelligent Algorithm

Fiber Bragg Grating(FBG)sensors have considerable advantages,such as high sensitivity,high accuracy,immunity to electromagnetic,stable chemical properties,compact size and light weight FBG sensors are mainly used to monitor physical quantities such as temperature,strain,pressure,and vibration.As the application range of FBG sensors expanding,the requirement for its reusability and stability have also increased.FBG sensor’s performance has three parts: reusability,stability of demodulation system,and stability of demodulation algorithm.The performance of the FBG sensor system is improved by intelligent algorithm.In order to improve the reusability of FBG sensing system,an overlapping spectral demodulation method based on Manta Ray Foraging Optimization(MRFO)algorithm is proposed.For the purpose of performance comparison,MRFO,particle swarm optimization(PSO),marine predators algorithm(MPA)and long short time memory(LSTM)are used for simulation analysis in the case of different number of FBG and different degree of overlap.And it is proved that MRFO has advantages in real-time performance and parameter tuning.In order to solve problem of the stability of demodulation algorithm,five reasons such as: initial population distribution,parameter settings of MRFO algorithm,noise of FBG sensor signal,the number of FBG and the degree of overlap,which leading to the local optimal solution of MRFO algorithm are analyzed and verified.A differential evolution marine predators algorithm(DEMRFO)is proposed,the Tent chaotic map is used to optimize the initial population,and the differential evolution algorithm is used to optimize the somersault foraging strategy.The DEMRFO can be divide into non-redundant DEMRFO and redundant DEMRFO according to whether redundant individuals are generated.The results show that the DEMRFO algorithm solves the problem of falling into the local optimal solution,and the demodulation stability of redundant DEMRFO is improved more obviously than the non-redundant DEMRFO.The demodulation error of distorted spectrum overlapping by two FBG is 4.6074 pm.Aiming at the center wavelength shift problem in the long-term use of FBG sensor,in order to improve the stability of demodulation system,an FBG wavelength shift predict method based on extreme learning machine(ELM)is proposed.The method trains the neural network by constructing the mapping relationship between the center wavelength and the wavelength shift.It can realize the prediction and generalization of the wavelength shift.The results show that the predicted error of FBG wavelength shift is less than 1 pm and the generalized error of FBG wavelength shift is less than 2 pm,which provides a useful exploration for the field calibration of FBG sensor system.