Artificial neural networks in the calibration of nonlinear mechanical models

Paper reviews applications of artificial neural networks in model calibration.Neural network-based calibration strategies are classified into three groups.Identification strategies are compared on calibration of affinity hydration model.The most precise strategy uses an ANN-based surrogate of each response component.Principal component-based inverse mapping is the best for a repeated use on new data. Rapid development in numerical modelling of materials and the complexity of new models increase quickly together with their computational demands. Despite the growing performance of modern computers and clusters, calibration of such models from noisy experimental data remains a nontrivial and often computationally intensive task. Layered neural networks provide a robust and efficient technique for overcoming the time-consuming simulations of calibrated models. The potential advantages of neural networks include simple implementation and high versatility in approximating nonlinear relationships. Therefore, there were several approaches proposed in literature for accelerating the calibration of nonlinear models by neural networks. This contribution reviews and compares three possible strategies based on approximating (i) the model response, (ii) the inverse relationship between the model response and its parameters and (iii) an error function quantifying how well the model fits the data. The advantages and drawbacks of particular strategies are demonstrated with the calibration of four parameters of an affinity hydration model from simulated data as well as from experimental measurements. The affinity hydration model is highly nonlinear but computationally cheap, thus allowing its calibration without any approximation and better quantification of results obtained by the examined calibration strategies. This paper can be viewed as a guide for engineers to help them develop an appropriate strategy for their particular calibration problems.

[1]  Anna Kucerová,et al.  Soft computing-based calibration of microplane M4 model parameters: Methodology and validation , 2013, Adv. Eng. Softw..

[2]  David Lehký,et al.  ANN inverse analysis based on stochastic small-sample training set simulation , 2006, Eng. Appl. Artif. Intell..

[3]  Habib N. Najm,et al.  Stochastic spectral methods for efficient Bayesian solution of inverse problems , 2005, J. Comput. Phys..

[4]  Patrick van der Smagt,et al.  Introduction to neural networks , 1995, The Lancet.

[5]  Luoxing Li,et al.  Inverse identification of interfacial heat transfer coefficient between the casting and metal mold using neural network , 2010 .

[6]  A. Kucerová.  Identification of nonlinear mechanical model parameters based on softcomputing methods , 2007 .

[7]  Zenon Waszczyszyn,et al.  Modal analysis and modified cascade neural networks in identification of geometrical parameters of circular arches , 2011 .

[8]  Michael T. Manry,et al.  An integrated growing-pruning method for feedforward network training , 2008, Neurocomputing.

[9]  Bernhard A. Schrefler,et al.  Hygro‐thermo‐chemo‐mechanical modelling of concrete at early ages and beyond. Part I: hydration and hygro‐thermal phenomena , 2006 .

[10]  Pietro Lura,et al.  Early development of properties in a cement paste: A numerical and experimental study , 2003 .

[11]  R. Baierlein.  Probability Theory: The Logic of Science , 2004 .

[12]  Hermann G. Matthies,et al.  Uncertainty Quantification with Stochastic Finite Elements , 2007 .

[13]  Joseph A. C. Delaney.  Sensitivity analysis , 2018, The African Continental Free Trade Area: Economic and Distributional Effects.

[14]  A. Ehrlacher,et al.  Analyses and models of the autogenous shrinkage of hardening cement paste: I. Modelling at macroscopic scale , 1995 .

[15]  Thomas J. R. Hughes,et al.  Encyclopedia of computational mechanics , 2004 .

[16]  Albert Tarantola,et al.  Inverse problem theory - and methods for model parameter estimation , 2004 .

[17]  G. Stavroulakis.  3.13 – Inverse Analysis , 2003 .

[18]  J. Shewchuk.  An Introduction to the Conjugate Gradient Method Without the Agonizing Pain , 1994 .

[19]  Hermann G. Matthies,et al.  Parameter Identification in a Probabilistic Setting , 2012, ArXiv.

[20]  E. Mizutani,et al.  Neuro-Fuzzy and Soft Computing-A Computational Approach to Learning and Machine Intelligence [Book Review] , 1997, IEEE Transactions on Automatic Control.

[21]  Ulrich Anders,et al.  Model selection in neural networks , 1999, Neural Networks.

[22]  John Moody,et al.  Prediction Risk and Architecture Selection for Neural Networks , 1994 .

[23]  Simon Haykin,et al.  Neural Networks and Learning Machines , 2010 .

[24]  Raphael T. Haftka,et al.  Surrogate-based Analysis and Optimization , 2005 .

[25]  Pavel Kordík,et al.  Meta-learning approach to neural network optimization , 2010, Neural Networks.

[26]  Anna Kucerová,et al.  Novel anisotropic continuum-discrete damage model capable of representing localized failure of massive structures. Part II: identification from tests under heterogeneous stress field , 2009, ArXiv.

[27]  Hedi Belhadjsalah,et al.  Parameter identification of an elasto-plastic behaviour using artificial neural networks–genetic algorithm method , 2011 .

[28]  Pavel Kordík.  GAME – Hybrid Self-Organizing Modeling System Based on GMDH , 2009 .

[29]  G. Liu,et al.  Rapid identification of elastic modulus of the interface tissue on dental implants surfaces using reduced-basis method and a neural network. , 2009, Journal of biomechanics.

[30]  Kurt Hornik,et al.  Some new results on neural network approximation , 1993, Neural Networks.

[31]  Anna Kucerová,et al.  Competitive Comparison of Optimal Designs of Experiments for Sampling-based Sensitivity Analysis , 2012, ArXiv.

[32]  Vít Smilauer,et al.  Fuzzy affinity hydration model , 2015, J. Intell. Fuzzy Syst..

[33]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[34]  Miguel Cervera,et al.  THERMO-CHEMO-MECHANICAL MODEL FOR CONCRETE. I: HYDRATION AND AGING , 1999 .

[35]  Roman Lackner,et al.  Back analysis of model parameters in geotechnical engineering by means of soft computing , 2003 .

[36]  Anna Kučerová,et al.  Acceleration of uncertainty updating in the description of transport processes in heterogeneous materials , 2011, J. Comput. Appl. Math..

[37]  Godfrey C. Onwubolu.  Hybrid Self-Organizing Modeling Systems , 2009 .

[38]  Ing Anička Kučerová.  Artificial Neural Networks in Calibration of Nonlinear Models , 2012 .

[39]  F. Cohen Tenoudji,et al.  Mechanical properties of cement pastes and mortars at early ages: Evolution with time and degree of hydration , 1996 .

[40]  Martin Abendroth,et al.  Identification of ductile damage and fracture parameters from the small punch test using neural networks , 2006 .

[41]  Tomáš Mareš,et al.  Application of Artificial Neural Networks in Identification of Affinity Hydration Model Parameters , 2012 .

[42]  Jon C. Helton,et al.  Survey of sampling-based methods for uncertainty and sensitivity analysis , 2006, Reliab. Eng. Syst. Saf..

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