1. Proceedings
  2. I3M
  3. 2018
  4. EMSS
  5. 10.46354/i3m.2018.emss.054

A multivariate model validation method based on kernel principal components analysis

  • a Yuchen Zhou   ,
  • b Ke Fang  ,
  • c Ping Ma  , 
  • d Ming Yang   
  • a, b, c, d  Control and Simulation Center, Harbin Institute of Technology
Cite as
Zhou Y., Fang K., Ma P., Yang M. (2018). A multivariate model validation method based on kernel principal components  analysis. Proceedings of the 30th European Modeling & Simulation Symposium (EMSS 2018), pp. 386-393. DOI: https://doi.org/10.46354/i3m.2018.emss.054

Abstract

Aimed at the similarity measure problem of multioutput simulation with nonlinear correlation between different output variables, a model validation method based on kernel principal component analysis (KPCA0 is proposed in this paper. KPCA is utilized to depict the relations among the multiple responses of observed data and simulation data respectively. A new similarity analysis formula is designed to measure the nonlinear
relations. The numerical experiment results reflect the effectiveness and reasonableness of the proposed validation method.

References

  1. Casas G.A., Rech P.C., 2012. Multistability annihilation in the Henon map through parameters modulation.
    Communication on Nonlinear Science and Numerical Simulation, 17 (6), 2570–2578
  2. Debruyne M., Verdonck T., 2010. Robust kernel principal component analysis and classification.
    Advances in Data Analysis and Classification, 4(2), 151–167.
  3. Fang K., Zhou Y.C., Zhao E.J., 2017. Discussion on the factor space of simulation model validation.
    Systems Engineering and Electronics, 39 (11), 2592–2602.
  4. Giannoulis C., Joakim J., Stromback P., Hellmans R., Heden H., 2013. Exercising GM-VV: verification
    and validation of a missile model. Proceedings of Fall Simulation Interoperability Workshop, pp.
    27–35. September 16–20, Orlando (USA).
  5. Guan H.S., Jiang Q.S., Wang S.R, 2009. Pattern matching method based on point distribution for
    multivariate time series. Journal of Software, 20(1), 67–79
  6. Jolliffe I.T., 2002. Principal Component Analysis, 2nd ed. New York: Springer
  7. Karamitopoulos L., Evangelidis G., Dervos D.A., 2008. Multivariate time series data mining: PCA-based measures for similarity search. Proceedings of the 4th International Conference on Data Mining, pp. 253–259. July 14–17, Las Vegas (USA
  8. Li W, Chen W., Jiang Z., Lu Z.Z., Liu Y., 2014. New validation metrics for models with multiple
    correlated responses. Reliability Engineering & System Safety, 127, 1–11.
  9. Mullins J., Ling Y., Mahadevan S., Sun L., Strachan A., 2016. Separation of aleatory and epistemic
    uncertainty in probabilistic model validation. Reliability Engineering & System Safety, 147, 49–
    59.
  10. Rebba R., Mahadevan S., 2006. Validation of models with multivariate output. Reliability Engineering
    & System Safety, 91 (8), 861–871.
  11. Roy C. J., Oberkampf W. L., 2011. A comprehensive framework for verification, validation, and uncertainty quantification in scientific Computer Methods in Applied Mechanics and Engineering, 200(25), 2131–2144.
  12. Sargent R.G. 2013. Verification and validation of simulation models. Journal of Simulation, 7 (1),
    12–24.
  13. Scholkopf B., Smola A., 2002. Learning with Kernels: Support Vector Machines, Regularization,
    Optimization, and Beyond. Cambridge: MIT Press.
  14. Scholkopf B., Smola A., Muller K., 1998. Nonlinear component analysis as a kernel eigenvalue
    problem. Neural Computation, 10 (5), 1299–1319.
  15. Wang H.Y, Lu S., 2006. Nonlinear Time Series Analysis and Its Application. Beijing: Science
    Press.
  16. Zhou Y.C., Fang K., Ma P., Yang M., 2018. Complex simulation model validation method based on
    ensemble learning. Systems Engineering and Electronics, 40, 1–9.