An adaptive fuzzy system for modeling chaos. (English) Zbl 0827.58039

Summary: Most chaotic systems do not have even dynamical behavior over the whole phase space. In some regions, the system stretches and branches more violently than others. In the regions where the dynamics are violent, finer representation must be given. In this paper, we make use of this property to model chaos. We present an adaptive system based on fuzzy logic. It can refine its representation of a region in the phase space if that region requires it. It does so by adaptively generating more fuzzy rules to model a region only if that region has very violent dynamics. Experiments were performed to test the adaptive fuzzy system for capturing the dynamics of a normal dynamical system (the Van der Pol oscillator) as well as two chaotic systems (the Lorenz and Rossler attractors). Results indicate that the fuzzy system can produce an accurate model of the three dynamical systems. The adaptive rule generation algorithm allowed the fuzzy system to have an optimal number of rules.


37D45 Strange attractors, chaotic dynamics of systems with hyperbolic behavior
93D05 Lyapunov and other classical stabilities (Lagrange, Poisson, \(L^p, l^p\), etc.) in control theory
03E72 Theory of fuzzy sets, etc.
Full Text: DOI


[1] Lapedes, A.; Farber, R., Nonlinear signal processing using neural networks: Prediction and system modelling, (Technical Report LA-UR-87-2662 (1987), Los Alamos Laboratory)
[2] Sato, M., A real time learning algorithm for recurrent analog neural networks, Biol. Cybernet., 62, 237-241 (1990) · Zbl 0686.92010
[3] Sato, M., A learning algorithm to teach spatiotemporal patterns to recurrent neural networks, Biol. Cybernet., 62, 259-263 (1990) · Zbl 0683.92006
[5] Pearlmutter, B. A., Dynamic recurrent neural networks, (Technical Report CMU-CS-90-196 (1990), School of Computer Science, Carnegie Mellon University)
[6] Williams, R. J.; Zipser, D., A learning algorithm for continually running fully recurrent neural networks, Neural Comput., 1, 270-280 (1989)
[7] Pearlmutter, B. A., Learning state space trajectories in recurrent neural networks, Neural Comput., 1, 263-269 (1989)
[8] Hiew, H. L., Learning chaotic behaviours using neural networks, (Honours Thesis (1991), Department of Computer Science, University of Western Australia)
[9] Kosko, B., Neural Networks and Fuzzy Systems: A Dynamical Systems Approach to Machine Learning (1992), Prentice-Hall: Prentice-Hall Englewood Cliffs, NJ · Zbl 0755.94024
[10] Henon, M., A two-dimensional mapping with a strange attractor, Commun. Math. Phys., 50, 69-77 (1976) · Zbl 0576.58018
[11] Packard, N. H.; Crutchfield, J. P.; Farmer, J. D.; Shaw, R. S., Geometry from a time series, Phys. Rev. Lett., 45, 712 (1980)
[12] Takens, F., Detecting strange attractors in fluid turbulence, (Rand, D.; Young, L. S., Dynamical Systems and Turbulence (1981), Springer: Springer Berlin) · Zbl 0513.58032
[13] Farmer, J. D.; Sidorowich, J. J., Predicting chaotic dynamics, (Kelso, J. A.S.; Mandell, A. J.; Shlesinger, M. F., Dynamic Patterns in Complex Systems (1988), World Scientific: World Scientific Singapore)
[14] Lorenz, E., Deterministic nonperiodic flow, J. Atmos. Sci., 20, 130 (1963) · Zbl 1417.37129
[15] Rossler, O. E., An equation for continuous chaos, Phys. Rev. Lett., 57A, 397-398 (1976) · Zbl 1371.37062
[16] Broomhead, D. S.; King, G. P., Extracting qualitative dynamics from experimental data, Physica, 20D, 217 (1987) · Zbl 0603.58040
[17] Farmer, J. D.; Sidorowich, J. J., Predicting chaotic time series, Phys. Rev. Lett., 59, 8, 845 (1987)
[18] Filev, D., Fuzzy modelling of complex systems, Int. J. Approx. Reasoning, 5, 281 (1991) · Zbl 0738.93048
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.