Model reduction technique for mechanical behaviour modelling: Efficiency criteria and validity domain assessment

The current paper presents the study of a neural network-based technique used to create fast, reduced, non-linear behavioural models. The studied approach is the use of artificial neural networks (ANNs) as a model reduction technique to create more efficient models, mostly in terms of computational speed. The test case is the deformation of a cantilever beam under large deflections (geometrical non-linearity). A reduced model is created by means of a multi-layer feed-forward neural network, a type of ANN reported as ‘universal approximator’ in the literature. Then it is compared with two finite-element models: linear (inaccurate for large deflections but fast) and non-linear (accurate but slow). Under large displacements, the reduced model approximates well the non-linear model while having similar speed to the linear model. Unfortunately, the resulting model presents a shortening of its validity domain, as being incapable of approximating the deformed configuration of the cantilever beam under small displacements. In other words, the ANN-based model provides a very good compromise between accuracy and speed within its validity domain, despite the low fidelity presented: accurate for large displacements but inaccurate for small displacements.

[1]  D. W. Scharpf,et al.  On the geometrical stiffness of a beam in space—a consistent V.W. approach , 1979 .

[2]  M. Lefik,et al.  Artificial neural network as an incremental non-linear constitutive model for a finite element code , 2003 .

[3]  Mark A. Ganter,et al.  Real-time finite element modeling for surgery simulation: an application to virtual suturing , 2004, IEEE Transactions on Visualization and Computer Graphics.

[4]  Sven Klinkel,et al.  Using finite strain 3D‐material models in beam and shell elements , 2002 .

[5]  R. Guyan.  Reduction of stiffness and mass matrices , 1965 .

[6]  P. Beran,et al.  Reduced-order modeling: new approaches for computational physics , 2004 .

[7]  J. N. Reddy,et al.  A corotational finite element formulation for the analysis of planar beams , 2005 .

[8]  Andrew J. Meade,et al.  The numerical solution of linear ordinary differential equations by feedforward neural networks , 1994 .

[9]  Wang Wei,et al.  On the estimation of the large deflection of a cantilever beam , 1993, Proceedings of IECON '93 - 19th Annual Conference of IEEE Industrial Electronics.

[10]  F. Filippou,et al.  Mixed formulation of nonlinear beam finite element , 1996 .

[11]  The cantilevered beam: an analytical solution for general deflections of linear-elastic materials , 2006 .

[12]  Michal Rewienski,et al.  A trajectory piecewise-linear approach to model order reduction of nonlinear dynamical systems , 2003 .

[13]  Dimitrios I. Fotiadis,et al.  Artificial neural networks for solving ordinary and partial differential equations , 1997, IEEE Trans. Neural Networks.

[14]  T. Low,et al.  The use of finite elements and neural networks for the solution of inverse electromagnetic problems , 1992, 1992. Digests of Intermag. International Magnetics Conference.

[15]  Igor N. Nikitin,et al.  Real-time simulation of elastic objects in Virtual Environments using finite element method and precomputed Green s functions , 2002, EGVE.

[16]  C. L. Philip Chen,et al.  A rapid supervised learning neural network for function interpolation and approximation , 1996, IEEE Trans. Neural Networks.

[17]  Kazufumi Kaneda,et al.  Direct solution method for finite element analysis using Hopfield neural network , 1995 .

[18]  Michel Tollenaere,et al.  Information structuring for use and reuse of mechanical analysis models in engineering design , 1999, J. Intell. Manuf..

[19]  Jens Kalkkuhl,et al.  FEM-based neural-network approach to nonlinear modeling with application to longitudinal vehicle dynamics control , 1999, IEEE Trans. Neural Networks.

[20]  Debasish Roy,et al.  Techniques based on genetic algorithms for large deflection analysis of beams , 2004 .

[21]  Fabrice Ville,et al.  Experimental and Numerical Investigations on the Air-Pumping Phenomenon in High-Speed Spur and Helical Gears , 2005 .

[22]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[23]  Christian Laugier,et al.  Physically realistic simulation of large deformations using LEM for interactive applications , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[24]  R. Taylor.  The Finite Element Method, the Basis , 2000 .

[25]  Giovanna Castellano,et al.  A neuro-fuzzy model reduction strategy , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[26]  K. Martini.  A PARTICLE-SYSTEM APPROACH TO REAL-TIME NON-LINEAR ANALYSIS , 2002 .

[27]  Todd K. Leen,et al.  Fast nonlinear dimension reduction , 1993, IEEE International Conference on Neural Networks.

[28]  Leonard Ziemiański,et al.  Neural networks in mechanics of structures and materials – new results and prospects of applications , 2001 .

[29]  Mingui Sun,et al.  Solving partial differential equations in real-time using artificial neural network signal processing as an alternative to finite-element analysis , 2003, International Conference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003.

[30]  Ralf Rabaetje,et al.  Real-time simulation of deformable objects for assembly simulations , 2003 .

[31]  George A. Hazelrigg,et al.  On the role and use of mathematical models in engineering design , 1999 .

[32]  Glaucio H. Paulino,et al.  Nonlinear Finite Element Analysis using an Object-Oriented Philosophy – Application to Beam Elements and to the Cosserat Continuum , 1999, Engineering with Computers.

[33]  Frédéric Boyer,et al.  Finite element of slender beams in finite transformations: a geometrically exact approach , 2004 .

[34]  L. C. Schmidt,et al.  Closed-form solution for the Timoshenko beam theory using a computer-based mathematical package , 1995 .

[35]  C. Neipp,et al.  Large and small deflections of a cantilever beam , 2002 .

[36]  R. V. Patel,et al.  A counter-propagation neural network for function approximation , 1990, 1990 IEEE International Conference on Systems, Man, and Cybernetics Conference Proceedings.

[37]  Qi Jing,et al.  Large-Deflection Beam Model for Schematic-Based Behavioral Simulation in NODAS , 2002 .

[38]  Mauro J. Atalla,et al.  Model updating using neural networks , 1996 .

[39]  Y. H. Kim,et al.  A predictor algorithm for fast geometrically-nonlinear dynamic analysis , 2003 .

[40]  Janko Slavič,et al.  Non-linearity and non-smoothness in multi-body dynamics: Application to woodpecker toy , 2006 .

[41]  Salvatore Coco,et al.  A multilayer perceptron neural model for the differentiation of Laplacian 3-D finite-element solutions , 2003 .

[42]  Shuixiang Li,et al.  Global flexibility simulation and element stiffness simulation in finite element analysis with neural network , 2000 .

[43]  Filip C. Filippou,et al.  Non-linear spatial Timoshenko beam element with curvature interpolation , 2001 .

[44]  Siak Piang Lim,et al.  A neural-network-based method of model reduction for the dynamic simulation of MEMS , 2001 .

[45]  Qingxin Yang,et al.  The use of neural networks combined with FEM to optimize the coil geometry and structure of transverse flux induction equipments , 2004 .

[46]  Geoffrey E. Hinton,et al.  NeuroAnimator: fast neural network emulation and control of physics-based models , 1998, SIGGRAPH.

[47]  Terence D. Sanger,et al.  A tree-structured adaptive network for function approximation in high-dimensional spaces , 1991, IEEE Trans. Neural Networks.

[48]  Michael A. Arbib,et al.  The handbook of brain theory and neural networks , 1995, A Bradford book.

[49]  Andrew J. Meade,et al.  Solution of nonlinear ordinary differential equations by feedforward neural networks , 1994 .

[50]  Gary R. Consolazio.  Iterative Equation Solver for Bridge Analysis Using Neural Networks , 2000 .

[51]  Guido Bugmann,et al.  NEURAL NETWORK DESIGN FOR ENGINEERING APPLICATIONS , 2001 .

[52]  Kin Keung Lai,et al.  An integrated data preparation scheme for neural network data analysis , 2006, IEEE Transactions on Knowledge and Data Engineering.

[53]  Alan J. Cartwright,et al.  Interactive prototyping—a challenge for computer based design , 1997 .

[54]  Daniel Svozil,et al.  Introduction to multi-layer feed-forward neural networks , 1997 .

[55]  Kuo Mo Hsiao,et al.  Nonlinear finite element analysis of elastic frames , 1987 .

[56]  J. N. Reddy,et al.  Unified finite elements based on the classical and shear deformation theories of beams and axisymmetric circular plates , 1997 .