Evolving fuzzy rule-based systems for modelling of non-linear non-stationary processes
Energy Saving Control, Volume # 1 | Part# 1
Digital Object Identifier (DOI)
fuzzy rule-based models,evolving fuzzy systems,on-line system identification,Kalman filter,recursive least squares,clustering
Recently introduced concept of evolving fuzzy modeling will be reviewed in this paper in the context of non-linear on-line modeling and identification of non-stationary processes. It will be presented as a logical consequence of the attempts to cope with the non-stationarity, non-linearity and the requirements for practical on-line algorithms that work in real-time and are close to the theoretically optimal/analytical solutions. The basic elements of the concept and its procedure will be outlined. This concept will be presented as a higher level adaptation that concerns model structure as well as model parameters. It will be presented as an extension of the multi-model concept and of the on-line identification of fixed structure fuzzy rule-based models. The name 'evolving' will be justified by the features of the algorithm such as 'inheritance', 'gradual change', 'learning by experience', 'self-organization', and 'age' that are typical for the evolution of living individuals and especially humans. One should distinguish this from the 'evolutionary computation' algorithms also called 'genetic' algorithms where the evolution of population of individuals is mimicked and operations such as 'crossover', 'mutation', 'reproduction' and 'selection' are used. From a pragmatic point of view, the new paradigm of 'evolving systems' is closer to on-line, real-time identification and adaptive systems branches of control theory while 'evolutionary algorithms' are heuristics which is firmly in the field of artificial intelligence. A number of application of this technique to a range of industrial and benchmark processes has been recently reported. Due to the lack of space only one problem related to building energy will be presented, namely modeling a typical air-conditioning unit will be presented.
 Kailath T., A. H. Sayed, B. Hassibi, Linear Estimation, Upper Saddle Rive, NJ, USA: Prentice Hall, 2000.  Ljung L. (1987) System Identification: Theory for the User, New Jersey, USA: Prentice-Hall.  Haykin, S., B. Widrow, Eds., Least-Mean-Squares Filters, New York, USA: John Willey and Sons, 2003.  Kalman R. E., A New Approach to linear filtering and prediction problem, Transactions of the ASME, Ser. D, Journal of Basic Engineering, vol. 82, pp. 34-45, 1960.  Haykin S., Adaptive Filter Theory, 4th ed. Upper Saddle River NJ, USA, Prentice Hall, 2002.  Pearson J., R. R. Goodal, M. Eastham, C. MacLeod, Investigation of Kalman Filter divergence using robust stability techniques, 36th Conference on Decision and Control, San Diego, CA, USA, pp. 4892-4893, 1997.  Chen R., J. S. Liu, Mixture Kalman Filters, Journal of the Royal Statistical Society, ser. B, vol. 62, pp. 493-508, 2000.  Crisan D., Particle Filters - A Theoretical perspective, Sequential Monte Carlo Methods in Practice, A Doucet, J. F. G., de freitas, N. J. Gordon, Eds., Berlin: Springer Verlag, 2001.  Murray-Smith, R., T. Johansen (Eds.), Multiple Model Approaches to Modelling and Control, Taylor and Francis, 1997.  Takagi, T., M. Sugeno, "Fuzzy identification of systems and its application to modeling and control", IEEE Trans. on Syst., Man & Cybernetics, vol. 15, pp. 116-132, 1985.  Fayyad, U. M., G. Piatetsky-Shapiro, P. Smyth, From Data Mining to Knowledge Discovery: An Overview, Advances in Knowledge Discovery and Data Mining, MIT Press. 1996.  Hornby A. S., Oxford Advance Learner's Dictionary, Oxford University Press, 1974.  Kasabov, N., "Evolving fuzzy neural networks for on-line supervised/unsupervised, knowledge-based learning," IEEE Trans. SMC - part B, Cybernetics 31, 902-918, 2001.  Angelov P., Evolving Rule-based Models: A Tool for Design of Flexible Adaptive Systems. Heidelberg, Germany: Springer-Verlag, 2002.  Michalewicz, Z., Genetic Algorithms + Data Structures = Evolution Programs, Berlin, Germany: Springer Verlag, 1996.  Koza J., Genetic Programming: On the Programming of Computers by Means of Natural Selection, USA: MIT Press, 1992.  Wang, L.-X., "Fuzzy Systems are Universal Approximators," Proc. FUZZ-IEEE, San Diego, CA, USA, pp. 1163-1170, 1992.  Yager R., D. Filev, Essentials of Fuzzy Modeling and Control, New York, USA: John Wiley and Sons, 1994.  F. Hoppner, F. Klawonn, R. Kruse, T. Runkler, Fuzzy Cluster Analysis, Chichester, UK: John Willey and Sons, 1999.  Kasabov N., Q. Song, "DENFIS: Dynamic Evolving Neural-Fuzzy Inference System and Its Application for Time-Series Prediction," IEEE Trans. on Fuzzy Systems, Vol. 10 (2), pp. 144-154, 2002.  Kim K., J. Baek, E. Kim, M. Park, "TSK Fuzzy model based on-line identification," Proc. 11th IFSA World Congress, Beijing, China, 2005, pp. 1435-1439.  Angelov, P., D. Filev, "An approach to on-line identification of evolving Takagi-Sugeno models", IEEE Trans. on Systems, Man and Cybernetics, part B, vol. 34, No. 1, pp. 484-498, 2004.  Leng G., T. M. McGuinty, G. Prasad, "An approach for on-line extraction of fuzzy rules using a self-organizing fuzzy neural network," Fuzzy Sets and Systems, vol. 150 (2), pp. 211-243, 2005.  Angelov, P., D. Filev, "Simpl_eTS: A Simplified Method for Learning Evolving Takagi-Sugeno Fuzzy Models," The 2005 IEEE Intern. Conf. on Fuzzy Systems FUZZ-IEEE, Reno, NE, USA, 2005, pp. 1068-1073.  Huang G.-B., P. Saratchandran, N. Sundarajan, "A generalized growing and pruning RBF (GGAP-RBF) neural network for function approximation," IEEE Trans. on NN, vol. 16 (1) 57-67, 2005.  Jang C.-F., X.-T. Lin, "A recurrent self-organizing neural fuzzy inference network," IEEE Trans. on Neural Networks, vol. 10, pp. 828-845, 1999.  Lin, F.-J., C.-H. Lin, P.-H. Shen, "Self-constructing fuzzy neural network speed controller for permanent-magnet synchronous motor drive," IEEE Trans. on Fuzzy Systems, Vol. 9 (5), pp. 751-759, 2001.  P. Angelov, D. Filev, N. Kasabov, O. Cordon (Eds.), Evovling Fuzzy Systems, Proc. 2nd International Symposium on Evolving Fuzzy Systems, 7-9 Sept. 2006, Ambelside, Lake District, UK, IEEE, 1-350 pp., to appear.  Angelov P., An Approach for Fuzzy Rule-base Adaptation using On-line Clustering, International Journal of Approximate Reasoning, Vol. 35 (3), March 2004, pp. 275-289.  Buswell, R. A., "Uncertainty in the First Principle Model based Condition Monitoring of HVAC Systems". Loughborough University, UK, PhD Thesis, 2001.