Visual reconstruction of flexible structure based on fiber grating sensor array and extreme learning machine algorithm

Abstract

A visual reconstruction method was proposed based on fiber Bragg grating (FBG) sensors and an intelligent algorithm, aiming to solve the problems of low accuracy and complex reconstruction process in conventional reconstruction methods of flexible structures. Firstly, the wavelength data containing structural strain information was captured by FBG sensors, together with deformation displacement information. Subsequently, a predicted model was built based on an extreme learning machine (ELM) and further optimized by the particle swarm optimization (PSO) algorithm. Different deformation patterns were tested on an aluminum alloy plate, indicating the ability of the predicted model to produce the deformation displacement for reconstruction. The experimental results show that the maximum error can be as low as 0.050 mm, which verifies that the proposed method is feasible and satisfied with the deformation monitoring of the spacecraft structure.

References

[1] GeJ, JamesA E, XuL, et al.. Bidirectional soft silicone curvature sensor based on off-centered embedded fiber Bragg grating[J]. IEEE photonics technology letters, 2016, 28(20):2237-2240

[2] SunG, LiH, DongM, et al.. Optical fiber shape sensing of polyimide skin for a flexible morphing wing[J]. Applied optics, 2017, 56(33):9325-9332

[3] TatabeK, IshikawaK, OikawaY. Compensation of fringe distortion for phase-shifting three-dimensional shape measurement by inverse map estimation[J]. Applied optics, 2016, 55(22):6017-6024

[4] WeiY, GaoF. Architecture design method for structural health monitoring system (SHM) of civil air-craft[C], 2017, InternationalConferenceonSensingDiagsticsProgsticsandControl: 736-739

[5] BangH J, JangM, ShinH, et al.. Structural health monitoring of wind turbines using fiber Bragg grating based sensing system[J]. Proc. SPIE, 2011, 7981: 79812H–79812H-8

[6] SahotaJ K, GuptaN, DhawanD. Fiber Bragg grating sensors for monitoring of physical parameters: a comprehensive review[J]. Optical engineering, 2020, 59(6):1

[7] ZengJ, WangW, WangB, et al.. Sensitivity of optical FBG sensor under dynamic/static load[J]. Journal of Nanjing University of Aeronautics & Astronautics, 2015, 47(03): 397-402

[8] ZhuX, LuM, FanH, et al.. Experimental research on intelligent structure vibration shape perception and reconstruction based on fiber grating network[J]. Chinese journal of scientific instrument, 2009, 30(001):65-70

[9] ZhangH, ZhuX, LiL, et al.. Space curved surface reconstruction method using two-dimensional curvature data[J]. Journal of basic science and engineering, 2015, 23(05):1035-1046

[10] WangC. 3D shapes detecting method of soft manipulator based on fiber Bragg grating sensor[J]. Control and instruments in chemical industry, 2015, 42(010):1130-1133

[11] ZhengJ, QianJ, ShenL, et al.. Surface reconstruction of 3D curved surface based on information of spatial curvature[J]. Journal of Shanghai University, 2009, 15(03):235-237250

[12] ThomasJ, GurusamyS, RajannaT R, et al.. Structural shape estimation by mode shapes using fiber Bragg grating sensors: a genetic algorithm approach[J]. IEEE sensors journal, 2020, 20(6):2945-2952

[13] BangH J, KoS W, JangM S, et al.. Shape estimation and health monitoring of wind turbine tower using a FBG sensor array[C], 2012, New York, IEEE: 496-500

[14] TesslerA, SpanglerJ L. A least-squares variational method for full-field reconstruction of elastic deformations in shear-deformable plates and shells[J]. Computer methods in applied mechanics & engineering, 2005, 194(2–5):327-339

[15] MaoZ, ToddM. Comparison of shape reconstruction strategies in a complex flexible structure[J]. Proc. SPIE, 2008, 6932: 69320H–69320H-12

[16] MishraM, SrivastavaM. A view of artificial neural network[C], 2014, New York, IEEE: 1-3

[17] HuangG B, ZhuQ Y, SiewC K. Extreme learning machine: a new learning scheme of feedforward neural networks[C], 2004, New York, IEEE: 985-990

[18] WangY, CaoF, YuanY. A study on effectiveness of extreme learning machine[J]. Neurocomputing, 2011, 74(16):2483-2490

[19] DingS, XuX, NieR. Extreme learning machine and its applications[J]. Neural computing and applications, 2013, 25(3–4):549-556

[20] WanC, XuZ, PinsonP, et al.. Optimal prediction intervals of wind power generation[J]. IEEE transactions on power systems, 2014, 29(3):1166-1174

[21] YeT, JianM, ChenL, et al.. Rolling bearing fault diagnosis under variable conditions using LMD-SVD and extreme learning machine[J]. Mechanism and machine theory, 2015, 90: 175-186

[22] KennedyJ, EberhartR. Particle swarm optimization[C], 1995, New York, IEEE: 1942-1948

[23] TreleaI C. The particle swarm optimization algorithm: convergence analysis and parameter selection[J]. Information processing letters, 2003, 85(6):317-325

[24] WangD, MengL. Performance analysis and parameter selection of PSO algorithms[J]. Acta automatica sinica, 2016, 42(010):1552-1561

[25] KinetD, MégretP, GoossenK W, et al.. Fiber Bragg grating sensors toward structural health monitoring in composite materials: challenges and solutions[J]. Sensors (Basel, Switzerland), 2014, 14(4):7394-7419

[26] HillK O, MeltzG. Fiber Bragg grating technology fundamentals and overview[J]. Journal of lightwave technology, 1997, 15(8):1263-1276

[27] Alvarez-BoteroG, BaronF E, CanoC C, et al.. Optical sensing using fiber Bragg gratings: fundamentals and applications[J]. IEEE instrumentation & measurement magazine, 2017, 20(2):33-38

[28] ZhangS, ChenM, HeQ, et al.. Quasidistributed fiber Bragg grating sensor network based on self-heterodyne detection technique[J]. Optical engineering, 2014, 53(5): 057107