Ann-based metamodelling with clustering of output values as an approach to robust inverse analysis

by

Ann-based metamodelling with clustering of output values as an approach to robust inverse analysis

Grzegorz Gorecki, Łukasz Rauch, Maciej Pietrzyk, Jan Kusiak

AGH University of Science and Technology, Department of Applied Computer Science and Modelling, Al. Mickiewicza 30, Krakow, Poland.

DOI:

https://doi.org/10.7494/cmms.2014.3.0488

Abstract:

Inverse analysis, which uses artificial neural networks as direct problem models, is gaining popularity. To improve robustness of the metamodel the idea of clustering of the networks based on their output was considered in the paper. This idea splits data by the output values, which in the inverse analysis are known as a part of the objective function. The individual networks are trained at small ranges of output values and they are supported by general networks trained on a wide range of data using special maps. The maps indicate which small network should be used to obtain more precise results. Possibility of using the general wide trained network for checking after the inverse analysis, whether the network for small range was correctly selected, is the main advantage of this method. The basic principles of this approach are described in the paper. Case study for identification of material flow stress model confirmed very good capabilities of this technique.

Cite as:

Gorecki, G., Rauch, Ł., Pietrzyk, M., & Kusiak, J. (2014). Ann-based metamodelling with clustering of output values as an approach to robust inverse analysis. Computer Methods in Materials Science, 14(3), 167 – 179. https://doi.org/10.7494/cmms.2014.3.0488

Article (PDF):

Keywords:

Inverse analysis, Artificial neural networks, Metamodel, Clustering by output values, Material flow stress

Publication dates:

Received: 31.07.2014, accepted: 12.11.2014, published:

Publication type:

Original scientific paper

References:

Ahmad, F., Matisa, N.A., Hussain, Z., Boudville, R., Osman, M.K., 2010, Genetic Algorithm – Artificial Neural Network (GA-ANN) hybrid intelligence for cancer diagno- sis, Int. Conf.e on Computational Intelligence, Liverpool, 78-83.

Bezdek, J. C., Tsao, E.C., Pal, N.K., 1992, Fuzzy Kohonen clustering networks, Proc. IEEE International Conference on Fuzzy System, San Diego, 1035-1043.

Forestier, R., Massoni, E., Chastel, Y., 2002, Estimation of constitutive parameters using an inverse method coupled to a 3D finite element software, Journal of Materials Processing Technology, 125, 594-601. // 1-s2.0- S0924013602004065-mainrrr.pdf

Gawąd, J., Kuziak, R., Madej, Ł., Szeliga, D., Pietrzyk, M., 2005, Identification of rheological parameters on the ba- sis of various types of compression and tension tests, Steel Research International, 76, 131-137.

Gavrus, A., Massoni, E., Chenot, J.L., 1996, An Inverse Analy- sis Using a Finite Element Model for Identification of Rheological Parameters, Journal of Materials Processing Technology, 60, 447-454.

Gelin, J.C., Ghouati, O., 1994, The inverse method for determining viscoplastic properties of aluminum alloys, Journal of Materials Processing Technology, 34, 435-440.

Hadasik, E., Kuziak, R., Kawalla, R., Adamczyk, M., Pietrzyk, M., 2006, Rheological model for simulation of hot rolling of new generation steel strips for automotive industry, Steel Research International, 77, 927-933.

Kermani, B.G., White, M.W., Nagle, H.T., 1995, Feature extraction by genetic algorithms for neural networks in breast cancer classification, Engineering in Medicine and Biology Society, IEEE 17th Annual Conference, 1, 831-832.

Khoddam, S., Lam, Y.C., Thomson, P.F., 1996, Numerical results achieved with an inverse computational method for determining the constitutive parameters using the hot torsion test results, Steel Research, 67, 39-43.

Kuncheva, L.I., 2005, Combining pattern classifiers, Methods and Algorithms, Wiley, New York.

Kusiak, J., Sztangret, Ł, Rauch, Ł., Pietrzyk, M., 2014, Meta- model driven optimization of thermomechanical indus- 14, 20-26.

Kuziak, R., Pietrzyk, M., 2011, Physical and numerical simula- tion of the manufacturing chain for the DP steel strips, Steel Research International, special edition conf. ICTP, 756-761.

Lazarevic, A., Obradovic, Z., 2001, Effective pruning of neural network classifier ensembles, Proc. Int. Joint Conf. on Neural Networks, 2, 796-801.

MacKay, D.J.C., 1992, Bayesian interpolation, Neural Computation, 4, 415-447.

Parvin, H., Alizadeh, H., Minaei-Bidgoli, B., Analoui, M., 2008, CCHR: Combination of classifiers using Heuristic Re- training, Int. Conf. on Networked Computing and advanced Information Management (NCM 2008), Gyeongju.

Pietrzyk, M., Kusiak, J., Kuziak, R., Zalecki, W., 2009, Optimization of laminar cooling of hot rolled DP steels, Proc. – 179 XXVIII Verformungskundliches Kolloquium, Plan-Neralm, 285-294.

Szeliga, D., Matuszyk, P., Kuziak, R., Pietrzyk, M., 2002, Identification of rheological parameters on the basis of various types of plastometric tests, Journal of Materials Processing Technology, 125-126, 150-154.

Szeliga, D., Gawąd, J., Pietrzyk, M., 2006, Inverse Analysis for Identification of Rheological and Friction Models in Metal Forming, Computer Methods in Applied Mechanics and Engineering, 195, 6778-6798.

Szeliga, D., Pietrzyk, M., 2012, Identification of rheological models and boundary conditions in metal forming, International Journal of Materials and Product Technology, 39, 388-405.

Szeliga, D., 2013, Identification problems in metal forming. A comprehensive study, Publisher AGH, 291, Kraków.

Sztangret, Ł., Milenin, A., Sztangret, M., Walczyk, W., Pie- trzyk, M., Kusiak, J., 2011a, Computer-aided design of the best TR forging technology for crank shafts, Computer Methods in Materials Science, 11, 237-242.

Sztangret, Ł., Szeliga, D., Kusiak, J., Pietrzyk, M., 2011b, Identification of material properties of the DP steel based on plastometric tests and on industrial hot strip rolling,

Sztangret, Ł., Szeliga, D., Kusiak, J., Pietrzyk, M., 2012, Application of the inverse analysis with metamodelling for the identification of the metal flow stress, Canadian Metallurgical Quarterly, 51, 440-446.

Tadeusiewicz R., Chaki R., Chaki N., 2014, Exploring Neural Networks with C#, CRC Press, Taylor & Francis Group, Boca Raton. Tadeusiewicz R., 2010, New Trends in Neurocybernetics.