Overcoming the Obstacle of Time-dependent Model Output for Statistical Analysis by Nonlinear Methods
Abstract
Modelica models represent static or dynamic systems. Their outputs can be scalar (numbers) or time-dependent (time series). The most advanced mathematical methods for the analysis of numerical models cannot cope with functional outputs. This paper aims to show an efficient method to reduce a time-dependent output to a few numbers. Principal component analysis is a well-established method for dimension reduction and can be used to tackle this issue. It relies, however, on a linear hypothesis that limits its applicability. We adapt and implement an existing method called the auto-associative model, invented by Stéphane Girard, to overcome this shortcoming. The auto-associative model generalizes PCA as it projects the data on a nonlinear (instead of linear) basis. It also provides physically interpretable data representations. The difference in efficiency between both methods is illustrated in a case study of the well-known bouncing ball model. We perform output reduction and reconstruction using both methods to compare the completeness of information kept throughout the dimension reduction process.
Doi: 10.28991/HIJ-2021-02-01-01
Full Text: PDF
Keywords
References
Fritzson, P. (2020). Modelica: Equation-Based, Object-Oriented Modelling of Physical Systems. Foundations of Multi-Paradigm Modelling for Cyber-Physical Systems, 45–96. doi:10.1007/978-3-030-43946-0_3.
Girard, S., & Yalamas, T. (2017). A probabilistic take on system modeling with modelica and python. Available online: https://tinyurl.com/proba-system-model. (accessed on January 2021).
Hotelling, H. (1933). Analysis of a complex of statistical variables into principal components. Journal of Educational Psychology, 24(6), 417–441. doi:10.1037/h0071325.
Houle, M. E. (2015). Inlierness, outlierness, hubness and discriminability: an extreme-value-theoretic foundation. National Institute of Informatics Technical Report NII-2015-002E, Tokyo, Japan.
Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., … Tarantola, S. (2007). Global Sensitivity Analysis. The Primer, Wiley Online Library. doi:10.1002/9780470725184.
Girard, S. (2014). Physical and Statistical Models for Steam Generator Clogging Diagnosis. Springer Briefs in Applied Sciences and Technology. doi:10.1007/978-3-319-09321-5.
Roustant, O., Ginsbourger, D., & Deville, Y. (2012). DiceKriging, DiceOptim: TwoRPackages for the Analysis of Computer Experiments by Kriging-Based Metamodeling and Optimization. Journal of Statistical Software, 51(1). doi:10.18637/jss.v051.i01.
Espadoto, M., Martins, R. M., Kerren, A., Hirata, N. S. T., & Telea, A. C. (2019). Towards a Quantitative Survey of Dimension Reduction Techniques. IEEE Transactions on Visualization and Computer Graphics, 1–1. doi:10.1109/tvcg.2019.2944182.
Sobol’, I. M. (1979). On the Systematic Search in a Hypercube. SIAM Journal on Numerical Analysis, 16(5), 790–793. doi:10.1137/0716058.
Girard, S. (2017). Otfmi: simulate FMU from OpenTURNS: User documentation. Technical Report RT-PMFRE-00997-003, Phimeca. Available online: https://github.com/openturns/otfmi. (accessed on January 2021).
Fukunaga, K., & Olsen, D. R. (1971). An Algorithm for Finding Intrinsic Dimensionality of Data. IEEE Transactions on Computers, C-20(2), 176–183. doi:10.1109/t-c.1971.223208.
Gerrer, C.-E., & Girard, S. (2019). Non Linear Dimension Reduction of Dynamic Model Output. Proceedings of the 13th International Modelica Conference, Regensburg, Germany, March 4–6, 2019. doi:10.3384/ecp19157189.
Girard, S., & Iovleff, S. (2008). Auto-Associative Models, Nonlinear Principal Component Analysis, Manifolds and Projection Pursuit. Principal Manifolds for Data Visualization and Dimension Reduction, 202–218. doi:10.1007/978-3-540-73750-6_8.
Sobol’, I. M. (2001). Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates. Mathematics and Computers in Simulation, 55(1-3), 271–280. doi:10.1016/s0378-4754(00)00270-6.
Campbell, K., McKay, M. D., & Williams, B. J. (2006). Sensitivity analysis when model outputs are functions. Reliability Engineering & System Safety, 91(10-11), 1468–1472. doi:10.1016/j.ress.2005.11.049.
Girard, S., Yalamas, T., & Baudin, M. (2018, October). Apprentissage statistique et modélisation 0D/1D des systèmes: application au vieillissement des batteries. Available online: https://hal.archives-ouvertes.fr/hal-02064908/ (accessed on January 2021).
Jolliffe, I. T., & Cadima, J. (2016). Principal component analysis: a review and recent developments. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 374(2065), 20150202. doi:10.1098/rsta.2015.0202.
M. M. Tiller, (2015) Modelica by Example, Xogeny. Available online: http://book.xogeny.com/behavior/discrete/bouncing.
Jansen, M. J. W. (1999). Analysis of variance designs for model output. Computer Physics Communications, 117(1-2), 35–43. doi:10.1016/s0010-4655(98)00154-4.
Saltelli, A., Annoni, P., Azzini, I., Campolongo, F., Ratto, M., & Tarantola, S. (2010). Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index. Computer Physics Communications, 181(2), 259–270. doi:10.1016/j.cpc.2009.09.018.
DOI: 10.28991/HIJ-2021-02-01-01
Refbacks
- There are currently no refbacks.
Copyright (c) 2020 Claire-Eleuthèriane Gerrer
