Case Study of Learning Entropy for Adaptive Novelty Detection in Solid-Fuel Combustion Control

  • Ivo BukovskyEmail author
  • Cyril Oswald
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 348)


This paper deals with the case study of usability of the Learning Entropy approach for the adaptive novelty detection in MIMO dynamical systems. The novelty detection is studied for typical parameters of linear systems including time delay. The solid-fuel combustion process is selected as a representative of typical non-linear dynamic MIMO system. The complex mathematical model of a biomass-fired 100kW boiler is used for verification of the potentials of the proposed method, and the motivation for novelty detection in solid-fuel combustion processes is discussed in this paper.


adaptive systems novelty detection learning entropy combustion process multiple input control incremental learning quadratic neural unit 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Markou, M., Singh, S.: Novelty detection: a review—part 1: statistical approaches. Signal Process. 83, 2481–2497 (2003), doi:10.1016/j.sigpro.2003.07.018CrossRefzbMATHGoogle Scholar
  2. 2.
    Markou, M., Singh, S.: Novelty detection: a review—part 2: neural network based approaches. Signal Process. 83, 2499–2521 (2003), doi:10.1016/j.sigpro.2003.07.019CrossRefzbMATHGoogle Scholar
  3. 3.
    Pincus, S.M.: Approximate entropy as a measure of system complexity. Proc. Natl. Acad. Sci. U S A 88, 2297–2301 (1991)CrossRefzbMATHMathSciNetGoogle Scholar
  4. 4.
    Richman, J.S., Moorman, J.R.: Physiological time-series analysis using approximate entropy and sample entropy. Am. J. Physiol. Heart Circ. Physiol. 278, H2039–H2049 (2000)Google Scholar
  5. 5.
    Marsland, S.: Novelty detection in learning systems. Neural Comput. Surv. 3, 157–195 (2003)Google Scholar
  6. 6.
    Demetriou, M.A., Polycarpou, M.M.: Incipient fault diagnosis of dynamical systems using online approximators. IEEE Trans. Autom. Control 43, 1612–1617 (1998), doi:10.1109/9.728881CrossRefzbMATHMathSciNetGoogle Scholar
  7. 7.
    Trunov, A.B., Polycarpou, M.M.: Automated fault diagnosis in nonlinear multivariable systems using a learning methodology. IEEE Trans. Neural Netw. 11, 91–101 (2000), doi:10.1109/72.822513CrossRefGoogle Scholar
  8. 8.
    Alippi, C., Roveri, M.: Just-in-Time Adaptive Classifiers #x2014;Part I: Detecting Nonstationary Changes. IEEE Trans. Neural Netw. 19, 1145–1153 (2008), doi:10.1109/TNN.2008.2000082CrossRefGoogle Scholar
  9. 9.
    Alippi, C., Bu, L., Zhao, D.: A prior-free encode-decode change detection test to inspect datastreams for concept drift. In: 2013 Int. Jt. Conf. Neural Netw., IJCNN, pp. 1–6 (2013)Google Scholar
  10. 10.
    Elwell, R., Polikar, R.: Incremental Learning of Concept Drift in Nonstationary Environments. IEEE Trans. Neural Netw. 22, 1517–1531 (2011), doi:10.1109/TNN.2011.2160459CrossRefGoogle Scholar
  11. 11.
    Bukovsky, I.: Learning Entropy: Multiscale Measure for Incremental Learning. Entropy 15, 4159–4187 (2013), doi:10.3390/e15104159CrossRefMathSciNetGoogle Scholar
  12. 12.
    Bukovsky, I., Homma, N., Cejnek, M., Ichiji, K.: Study of Learning Entropy for Novelty Detection in lung tumor motion prediction for target tracking radiation therapy. In: 2014 Int. Jt. Conf. Neural Netw., IJCNN, pp. 3124–3129 (2014)Google Scholar
  13. 13.
    Bukovsky, I., Oswald, C., Cejnek, M., Benes, P.M.: Learning entropy for novelty detection a cognitive approach for adaptive filters. Sens. Signal Process. Def. SSPD 2014, pp. 1–5 (2014)Google Scholar
  14. 14.
    Bukovsky, I., Kinsner, W., Bila, J.: Multiscale analysis approach for novelty detection in adaptation plot. Institution of Engineering and Technology, pp. 27–27 (2012)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Department of Instrumentation and Control EngineeringCzech Technical University in PraguePragueCzech Republic

Personalised recommendations