Impact load identification of nonlinear structures using deep Recurrent Neural Network

Abstract In this paper, a novel impact load identification method of nonlinear structures by using deep Recurrent Neural Network (RNN) is proposed. The deep RNN model, mainly consisting of two Long Short-Term Memory (LSTM) layers and one bidirectional LSTM (BLSTM) layer, is trained through a large number of dynamic responses and impact loads to learn the complex inverse mapping between structural inputs and outputs. The effectiveness and practicability of the proposed method are verified by three nonlinear cases: damped Duffing oscillator, nonlinear three-degree-of-freedom system and nonlinear composite plate. The results show that the proposed method has the capability for identifying the complex impact load even when the impact location is unknown. Meanwhile, hyperparameters of the deep RNN model and placement scheme of sensors are not highly sensitive to the identification accuracy.

[1]  F. Gunawan,et al.  Impact-Force Estimation by Quadratic Spline Approximation , 2008 .

[2]  Hirotsugu Inoue,et al.  Inverse analysis of the magnitude and direction of impact force , 1995 .

[3]  Mohammad I. Albakri,et al.  Impact localization in dispersive waveguides based on energy-attenuation of waves with the traveled distance , 2018 .

[4]  Abdellatif Khamlichi,et al.  Assessing impact force localization by using a particle swarm optimization algorithm , 2014 .

[5]  Keith Worden,et al.  Impact detection in an aircraft composite panel—A neural-network approach , 2007 .

[6]  Zubaidah Ismail,et al.  Impact Force Identification With Pseudo-Inverse Method on a Lightweight Structure for Under-Determined, Even-Determined and Over-Determined Cases , 2014 .

[7]  Tommy H.T. Chan,et al.  Moving force identification based on modified preconditioned conjugate gradient method , 2018, Journal of Sound and Vibration.

[8]  Li Zhou,et al.  Impact load identification of composite structure using genetic algorithms , 2009 .

[9]  Z. Sharif-Khodaei,et al.  Identification of impact force for smart composite stiffened panels , 2013 .

[10]  Hisao Fukunaga,et al.  An efficient approach for identifying impact force using embedded piezoelectric sensors , 2007 .

[11]  Zahra Sharif Khodaei,et al.  An Event-Triggered Energy-Efficient Wireless Structural Health Monitoring System for Impact Detection in Composite Airframes , 2019, IEEE Internet of Things Journal.

[12]  Ruonan Liu,et al.  The application of cubic B-spline collocation method in impact force identification , 2015 .

[13]  Anthony N. Palazotto,et al.  Scaling numerical models for hypervelocity test sled slipper-rail impacts , 2006 .

[14]  Yongan Huang,et al.  Impact Monitoring for Aircraft Smart Composite Skins Based on a Lightweight Sensor Network and Characteristic Digital Sequences , 2018, Sensors.

[15]  Xuefeng Chen,et al.  Sparse regularization for force identification using dictionaries , 2016 .

[16]  Yann LeCun,et al.  The Loss Surfaces of Multilayer Networks , 2014, AISTATS.

[17]  W. Krenkel,et al.  C/C–SiC composites for space applications and advanced friction systems , 2005 .

[18]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[19]  Taek Soo Jang,et al.  A new method for measuring nonharmonic periodic excitation forces in nonlinear damped systems , 2011 .

[20]  Keith Worden,et al.  Fail-safe sensor distributions for impact detection in composite materials , 2000 .

[21]  Jan Holnicki-Szulc,et al.  On-line impact load identification , 2013 .

[22]  M. H. Aliabadi,et al.  Reliability based impact localization in composite panels using Bayesian updating and the Kalman filter , 2018 .

[23]  Fergyanto E. Gunawan,et al.  Levenberg-Marquardt iterative regularization for the pulse-type impact-force reconstruction , 2012 .

[24]  Jürgen Schmidhuber,et al.  LSTM: A Search Space Odyssey , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[25]  Taek Soo Jang,et al.  A method for simultaneous identification of the full nonlinear damping and the phase shift and amplitude of the external harmonic excitation in a forced nonlinear oscillator , 2013 .

[26]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[27]  Keith Worden,et al.  An automatic impact monitor for a composite panel employing smart sensor technology , 2005 .

[28]  Baijie Qiao,et al.  Sparse deconvolution for the large-scale ill-posed inverse problem of impact force reconstruction , 2017 .

[29]  Jang-Kyo Kim,et al.  Vibration damping characteristics of carbon fiber-reinforced composites containing multi-walled carbon nanotubes , 2011 .

[30]  Sang-Kwon Lee,et al.  Experimental identification for inverse problem of a mechanical system with a non-minimum phase based on singular value decomposition , 2008 .

[31]  M. H. Ferri Aliabadi,et al.  Passive sensing method for impact localisation in composite plates under simulated environmental and operational conditions , 2019, Mechanical Systems and Signal Processing.

[32]  Keqiang Li,et al.  Technical note: Coherence analysis of the transfer function for dynamic force identification , 2011 .

[33]  Shilin Xie,et al.  Identification of high frequency loads using statistical energy analysis method , 2013 .

[34]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[35]  M. T. Paridah,et al.  A review on dynamic mechanical properties of natural fibre reinforced polymer composites , 2016 .

[36]  Jae-Hung Han,et al.  Separation characteristics study of ridge-cut explosive bolts , 2014 .

[37]  Mark M. Derriso,et al.  Impact Loads Identification in Standoff Metallic Thermal Protection System Panels , 2007 .

[38]  Minglong Xu,et al.  Simulated and experimental studies on identification of impact load with the transient statistical energy analysis method , 2014 .

[39]  R. Hashemi,et al.  Vibration Base Identification of Impact Force Using Genetic Algorithm , 2007 .

[40]  Xu Guo,et al.  A state space force identification method based on Markov parameters precise computation and regularization technique , 2010 .

[41]  Taek Soo Jang,et al.  Indirect measurement of the impulsive load to a nonlinear system from dynamic responses: Inverse problem formulation , 2010 .

[42]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[43]  E. Jacquelin,et al.  Force reconstruction: analysis and regularization of a deconvolution problem , 2003 .

[44]  William P. Schonberg,et al.  Protecting Earth-orbiting spacecraft against micro-meteoroid/orbital debris impact damage using composite structural systems and materials: An overview , 2010 .

[45]  Qingshan Yang,et al.  Sensor placement methods for an improved force identification in state space , 2013 .

[46]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[47]  Chih-Kao Ma,et al.  An inverse method for the estimation of input forces acting on non-linear structural systems , 2004 .

[48]  Bor-Tsuen Wang,et al.  DETERMINATION OF UNKNOWN IMPACT FORCE ACTING ON A SIMPLY SUPPORTED BEAM , 2003 .

[49]  Wang Zhen,et al.  A dynamic load estimation method for nonlinear structures with unscented Kalman filter , 2018 .