Prognostics for Aircraft Control Surface Damage Based on Fuzzy Least Squares Support Vector Regression (FLS-SVR)

  • Lei Dong
  • Zhang Ren
  • Qingdong Li
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 324)


The trends of flight control system state parameters which can be measured are indirect manifestations of surface damage. In order to predict the changes of states trend more accurately, an algorithm based on fuzzy least squares support vector regression (FLS-SVR) was presented. This approach reconstructed the phase space of multivariate time series using K-L transformation method. A FLS-SVR model was built with the new information priority theory according to apply a fuzzy membership to each input point. The SVR parameters were optimized by genetic algorithm (GA) to improve the accuracy of the model. In order to verify the validity of FLS-SVR algorithm, the prognostics and analysis of surface damage trend were performed. The simulation result demonstrates the prognostics model has good predictive ability.


Prognostic support vector regression fuzzy membership least squares genetic algorithm 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Belkharraz, A.I., Sobel, K.: Direct adaptive control for aircraft control surface failures. In: Proceedings of the American Control Conference, pp. 3905–3910. IEEE, Piscataway (2003)Google Scholar
  2. 2.
    Weiss, J.L., Willsky, A.S., Looze, D.P., et al.: Detection and isolation of control surface effectiveness failures in high performance aircraft. In: IEEE NAECON, pp. 552–559 (1985)Google Scholar
  3. 3.
    Zhang, Y.Z., Deng, J.H.: On-Line fault mode prediction and fault detection for control surface damage. Journal of Northwestern Polytechnical University 21, 298–301 (2003)Google Scholar
  4. 4.
    Li, B., Zhang, W.G., Ning, D.F., Yin, W.: Fault prediction system of airplane steer surface based on neural network model. Journal of System Simulation 20, 5840–5842 (2008)Google Scholar
  5. 5.
    Rosenstein, M.T., Collins, J.J., De Luca, C.J.: Reconstruction expansion as a geometry based framework for choosing proper delay yime. Physica D: Nonlinear Phenomena 73, 82–98 (1994)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Schouten, J.C., Coppens, M.-O., Takens, F., van den Bleek, C.M.: Chaotic attractor learning and how to deal with nonlinear singularities. In: The IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium, AS-SPCC 2000, October l-4, pp. 466–470 (2000)Google Scholar
  7. 7.
    Srinivas, J., Catnaik, L.M.: Adaptive probabilities of crossover and mutation in genetic algorithm. IEEE Transactions on SMC 24, 656–667 (1994)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Lei Dong
    • 1
  • Zhang Ren
    • 1
  • Qingdong Li
    • 1
  1. 1.Science and Technology on Aircraft Control LaboratoryBeihang UniversityBeijingChina

Personalised recommendations