Self-evolution of hyper fractional order chaos driven by a novel approach through genetic programming

https://doi.org/10.1016/j.eswa.2015.12.033Get rights and content

Highlights

  • Fractional order chaos (FOC) evolution.

  • United functional extrema model (UFEM).

  • Self-evolution for FOC.

  • Driven by Genetic programming through reconstruction.

Abstract

To find best inherent chaotic systems behind the complex phenomena is of vital important in Complexity science research. In this paper, a novel non-Lyapunov methodology is proposed to self-evolve the best hyper fractional order chaos automatically driven by a computational intelligent method, genetic programming. Rather than the unknown systematic parameters and fractional orders, the expressions of fractional-order differential equations (FODE) are taken as particular independent variables of a proper converted non-negative minimization of special functional extrema in the proposed united functional extrema model (UFEM), then it is free of the hypotheses that the definite forms of FODE are given but some parameters and fractional orders unknown. To demonstrate the potential of the proposed methodology, simulations are done to evolve a series of benchmark hyper and normal fractional chaotic systems in complexity science. The experiments’ results show that the proposed paradigm of fractional order chaos driven by genetic programming is a successful method for chaos’ automatic self-evolution, with the advantages of high precision and robustness.

Introduction

Complex systems are ubiquitous in physics, sociology, biology, economics and many other scientific areas (Holland, 1998). Normally, a complex system consists of tiny aggregated components, whose interaction and interconnectedness are nontrivial anyway. Meanwhile, the interactions result in high-dimensional, nonlinear interactions and connectivity leads to nontrivial topological features such as high clustering coefficients and power-law degree distribution. These bring about emergent properties of the complex system, not anticipated by its own isolated components. Furthermore, when system behaviour is studied form a temporal perspective, self-organisation patterns come into being typically. After the symbolic event that Mandelbort discovered that there were a lot of fractional dimension phenomena in nature (Mandelbrot, 1982) in bifurcation, hyperchaos, proper and improper fractional-order chaos systems and chaos synchronization (Diethelm, Ford, 2002, Doha, Bhrawy, Ezz-Eldien, 2012, Odibat, Corson, Aziz-Alaoui, Bertelle, 2010, Si, Sun, Zhang, Chen, 2012b, Song, Yang, Xu, 2010, Tavazoei, Haeri, 2007, Tavazoei, Haeri, 2008), the applications of fractional differential equations began to appeal to related scientists (Agrawal, Srivastava, Das, 2012, Bhalekar, Daftardar-Gejji, 2010, Diethelm, Ford, Freed, 2002, Grigorenko, Grigorenko, 2003, Kaslik, Sivasundaram, 2012, Odibat, 2012, Odibat, Corson, Aziz-Alaoui, Bertelle, 2010, Tang, Zhang, Hua, Li, Yang, 2012, Tavazoei, Haeri, 2007).

Studying complex fractional order systems requires composite strategies, which employ diverse algorithms to solve a single but difficult problem. Components of these strategies may resolve consecutive phases leading to the final goal, or may resolve the main problem in different methods which can be aggregated to form the final solution. For example, the hyper-heuristics, evolutionary algorithms (Gao, Fei, Lee, Tong, Deng, Zhao, 2014a, Gao, Lee, Fei, Tong, Deng, Zhao, 2014b), island Genetic algorithms (GA), swarm intelligent methods (Gao, Lee, Tong, Fei, Zhao, 2013, Gao, Qi, Yin, Xiao, 2010a, Gao, Qi, Yin, Xiao, 2010b, Gao, Tong, 2006) and the other approaches are often used.

The paper will proposed a novel approach to simulate the evolution process driven by Genetic Programming (GP) of the fractional order complex systems in an generalized identification perspective. It is difficult to identify the expressions or parameters in the uncertain fractional-order chaotic systems. There are some systematic parameters and orders are unknown for the fractional-order chaos systems in controlling and synchronization.

Hitherto, there have been at two main approaches in generalized parameters’ identification for fractional-order chaos systems. The first is a Lyapunov way using synchronisation. And there have been few results on parameter estimation method of fractional-order chaotic systems based on chaos synchronization (Si, Sun, Zhang, & Zhang, 2012a). However, the design of controller and the updating law of parameter identification is a task with technique and sensitively depends on the considered systems. The second is a non-Lyapunov way using some artificial intelligence methods, such as differential evolution (Tang et al., 2012) and particle swarm optimization (Yuan & Yang, 2012).

Although there are many methods in parameters estimation for integer order chaos systems (Alonso, Henríquez, Henríquez, Rodríguez-Herrera, Bolaños, Alpízar, Travieso, Cabrera, 2015, Chang, Yang, Liao, Yan, 2008a, Fan, Li, Tian, 2015, Gao, xia Fei, fang Deng, bo Qi, Ilangko, 2012a, Gao, Qi, Balasingham, Yin, Gao, 2012b, Mukhopadhyay, Banerjee, 2012, Pan, Das, 2015, dos Santos, Luvizotto, Mariani, dos Santos Coelho, 2012, Wang, Xu, Li, 2011) , to the best of our knowledge, a little work in non-Lyapunov way has been done to the unknown expression or the parameters and orders estimation of fractional-order chaos systems (Gao, Fei, Lee, Tong, Deng, Zhao, 2014a, Gao, Lee, Fei, Tong, Deng, Zhao, 2014b, Gao, Lee, Tong, Fei, Zhao, 2013, Tang, Zhang, Hua, Li, Yang, 2012, Yuan, Yang, 2012).

For example, we consider the following fractional-order chaos system. αDtqY(t)=f(Y(t),Y0(t),θ)where Y(t)=(y1(t),y2(t),,yn(t))Tn denotes the state vector. θ=(θ1,θ2,,θn)T denotes the original parameters. q=(q1,q2,,qn),(0<qi<1,i=1,2,,n) is the fractional derivative orders. f(Y(t),Y0(t),θ)=(f1,f2,,fn)|(Y(t),Y0(t),θ)

The continuous integro-differential operator (Petráš, 2011a, Petráš, 2011b) is defined as αDytq={dqdxq,q>0;1,q=0;α1(dτ)q.

Normally the function expression f in Eq. (1) is known. But some the θ, q are unknown , then the Θ=(θ1,θ2,,θn, q1, q2,…, qn) will be the parameters to be estimated, through a constructed evaluation function F. And the objective is obtained as following Eq. (2), θ*=argminθF

However, there exist basic hypotheses in traditional non-Lyapunov estimation methods (Al-Assaf, El-Khazali, Ahmad, 2004, Tang, Zhang, Hua, Li, Yang, 2012, Yuan, Yang, 2012) on generalized parameters’ identification. That is, the parameters and fractional orders Θ=(θ1,θ2,,θn,q1,q2,,qn) are partially unknown to be estimated. But few works are valid in the case the definite expression forms f=(f1,f2,,fn). What should I do when these hypotheses do not exist? That is, when some the fractional order differential equations (FODE)’ expression f=(f1,f2,,fn) are unknown, how to identify the fractional system in general? That is, (f1,f2,,fn)*=argmin(f1,f2,,fn)F

Now the problem of parameters estimation (2) become another much more complicated question, to find the forms of fractional order equations as in (3). In another word, it is fractional-order chaos’ evolution driven by some methods in reconstruction perspective now.

However, to the best of authors’ knowledge, there are no methods in non-Lyapunov way for fractional order chaotic systems’ reconstruction for system evolution so far. The objective of this work is to present a novel simple but effective approach to evolve the hyper, proper and improper fractional chaotic systems in a non-Lyapunov way. In which, an idea of self-evolved fractional order chaos driven by genetic programming are used to identify unknown expressions’ forms of FODE f=(f1,f2,,fn), in which no extra evolutionary algorithms to be added or be compounded. And the illustrative evolution simulations for diverse chaos systems system are discussed respectively.

The rest is organized as follows. Section 2 gives a simple review on non-Lyapunove parameters estimation methods for fractional-order chaos systems. In Section 3, a novel united mathematical model for fractional chaos evolution are proposed in a non-Lyapunov way. Then all the existing non-Lyapunov methods in Section 2 become special cases of the new united model, such as non-Lyapunov parameters identification for normal and fractional chaos systems, non-lyapunov reconstruction methods for normal chaos system. In Section 3.3 a novel methods with proposed united model in Section 3 is introduced to reconstruct the hyper, improper and normal fractional chaos systems. And simulations are done to a series of different fractional chaos systems and their special cases in three dimension. Conclusions are summarized briefly in Section 5.

Section snippets

Non-Lyapunove reconstruction methods for normal chaos systems

To solve the question proposed in Section 1, a new ideas are introduced to generalize the normal parameters’ estimation problems into fractional order chaos reconstruction problems. In the sense of the differential equations for chaos system reconstruction, the parameters’ identification for normal chaos system cases (Chang, Yang, Liao, Yan, 2008b, Chang, 2007, Gao, Lee, Li, Tong, Lü, 2009b, Gao, Li, Tong, 2008, Gao, Qi, Balasingham, Yin, Gao, 2012b, Gao, Qi, Yin, Xiao, 2012c, Gao, Tong, 2006,

United functional extrema model (UFEM) for fractional chaos evolution

Unlike the normal Lyapunov methods (Ghorbani, Kisi, Aalinezhad, 2010, Lu, Lü, Xie, Chen, 2003, Toledo-Suárez, 2010) for the normal chaos systems, there have been some results on chaos reconstruction for f=(f1,f2,,fn) with the non-Lyapunove methods mainly from symbolic regression through genetic programming (GP) (Koza, 1992, Koza, 1994, Poli, Langdon, McPhee, 2008), and some evolutionary algorithms (Gandomi, Alavi, 2012a, Gandomi, Alavi, 2012b, Ikeda, 2003, Kirstukas, Bryden, Ashlock, 2005,

Simulations and discussions

To test the Algorithm 1, some different well known and widely used Hyper, proper and improper fractional chaos systems (Gao, Fei, Lee, Tong, Deng, Zhao, 2014a, Gao, Lee, Fei, Tong, Deng, Zhao, 2014b) are selected.

Conclusions

For the intelligent systems to discover the secrets, disciplines, the deep scientific principles of the nature or social complex phenomena (Holland, 1998), automatic finding the inherent representing these is a good way what we want to realize. The fractional chaotic systems are some of these inherent things.

In the paper, we proposed an automatic self-evolving paradigm to realize the above objective driven by GP theoretically. And the works (Alonso, Henríquez, Henríquez, Rodríguez-Herrera,

References (90)

  • GaoF. et al.

    Inversion mechanism with functional extrema model for identification incommensurate and hyper fractional chaos via differential evolution

    Expert Systems with Applications

    (2014)
  • GaoF. et al.

    Identification of unknown parameters and orders via cuckoo search oriented statistically by differential evolution for noncommensurate fractional-order chaotic systems

    Abstract and Applied Analysis

    (2013)
  • GaoF. et al.

    A novel non-lyapunov way for detecting uncertain parameters of chaos system with random noises

    Expert Systems with Applications

    (2012)
  • GaoF. et al.

    A novel non–lyapunov approach in discrete chaos system with rational fraction control by artificial bee colony algorithm

    Proceedings of the 2010 international conference on progress in informatics and computing(pic-2010)

    (2010)
  • GaoF. et al.

    An novel optimal pid tuning and on–line tuning based on artificial bee colony algorithm

    Proceedings of the 2010 international conference on computational intelligence and software engineering (cise 2010)

    (2010)
  • GaoF. et al.

    Online synchronization of uncertain chaotic systems by artificial bee colony algorithm in a non–lyapunov way

    Proceedings of the 2010 international conference on computational intelligence and software engineering (cise 2010)

    (2010)
  • GhorbaniM.A. et al.

    A probe into the chaotic nature of daily streamflow time series by correlation dimension and largest lyapunov methods

    Applied Mathematical Modelling

    (2010)
  • KilbasA.A. et al.

    Theory and applications of fractional differential equations

    (2006)
  • LiC. et al.

    On the fractional adams method

    Computers and Mathematics with Applications

    (2009)
  • OdibatZ.M. et al.

    Synchronization of chaotic fractional-order systems via linear control

    International Journal of Bifurcation and Chaos

    (2010)
  • O’NeillM. et al.

    Grammatical evolution: evolutionary automatic programming in an arbitrary language

    (2003)
  • ParlitzU.

    Estimating model parameters from time series by autosynchronization

    Physical Review Letters

    (1996)
  • PetrášI.

    Fractional calculus

    Fractional-order nonlinear systems: modeling, analysis and simulation

    (2011)
  • PoliR. et al.

    A field guide to genetic programming

    (2008)
  • dos SantosG.S. et al.

    Least squares support vector machines with tuning based on chaotic differential evolution approach applied to the identification of a thermal process

    Expert Systems with Applications

    (2012)
  • ShenL. et al.

    Robust synchronization and parameter identification on a class of uncertain chaotic systems

    Chaos, Solitons & Fractals

    (2008)
  • VaradanV. et al.

    Reconstruction of polynomial systems from noisy time-series measurements using genetic programming

    IEEE Transactions on Industrial Electronics

    (2001)
  • WuX.J. et al.

    Chaos in the fractional-order lorenz system

    International Journal of Computer Mathematics

    (2009)
  • YangK. et al.

    Parameters identification of chaotic systems by quantum-behaved particle swarm optimization

    International Journal of Computer Mathematics

    (2009)
  • YuY. et al.

    Dynamic analysis of a fractional-order lorenz chaotic system

    Chaos, Solitons and Fractals

    (2009)
  • YuanL.-G. et al.

    Parameter identification and synchronization of fractional-order chaotic systems

    Communications in Nonlinear Science and Numerical Simulation

    (2012)
  • ZelinkaI. et al.

    An investigation on evolutionary identification of continuous chaotic systems

    American institute of physics conference series

    (2011)
  • ZelinkaI. et al.

    An investigation on evolutionary reconstruction of continuous chaotic systems

    Mathematical and Computer Modelling

    (2013)
  • ZelinkaI. et al.

    Evolutionary reconstruction of chaotic systems

  • ZhangW. et al.

    Genetic programming-based chaotic time series modeling

    Journal of Zhejiang University Science

    (2004)
  • Al-AssafY. et al.

    Identification of fractional chaotic system parameters

    Chaos, Solitons &; Fractals

    (2004)
  • CastleT. et al.

    Positional effect of crossover and mutation in grammatical evolution

  • ChangJ.-F. et al.

    Parameter identification of chaotic systems using evolutionary programming approach

    Expert Systems with Applications

    (2008)
  • ChangJ.-F. et al.

    Parameter identification of chaotic systems using evolutionary programming approach

    Expert Systems with Applications

    (2008)
  • ChangW.-D.

    Parameter identification of chen and lu systems: a differential evolution approach

    Chaos Solitons & Fractals

    (2007)
  • DiethelmK.

    An efficient parallel algorithm for the numerical solution of fractional differential equations

    Fractional Calculus and Applied Analysis

    (2011)
  • FanX. et al.

    Chaotic characteristic identification for carbon price and an multi-layer perceptron network prediction model

    Expert Systems with Applications

    (2015)
  • GandomiA. et al.

    A new multi-gene genetic programming approach to nonlinear system modeling. part i: materials and structural engineering problems

    Neural Computing & Applications

    (2012)
  • GaoF. et al.

    Detecting unstable periodic orbits of nonlinear mappings by a novel quantum-behaved particle swarm optimization non-lyapunov way

    Chaos, Solitons & Fractals

    (2009)
  • GaoF. et al.

    Parameter estimation for chaotic system with initial random noises by particle swarm optimization

    Chaos, Solitons & Fractals

    (2009)
  • Cited by (5)

    • Evolving chaos: Identifying new attractors of the generalised Lorenz family

      2018, Applied Mathematical Modelling
      Citation Excerpt :

      With such an aim, even sub-optimal intermediate solutions with slightly positive Lyapunov exponents may show rich phase space dynamics as the complexity does not necessarily always correlate with high LLE. This is the reason why we here report even the intermediate search results with LLE > 0, similar to the earlier explorations in [17], during the evolutionary search process and not only at the final converged results of GP algorithm, which has been misinterpreted in Gao et al. [31]. Also, in this paper, we have listed the newly found chaotic system as the GP search progressed and not using the calculated LLE.

    • Transmission and Decryption of the Audio Signal Masked with ECG by FDM Method

      2022, Iranian Journal of Science and Technology - Transactions of Electrical Engineering

    The work was carried out during the tenure of the ERCIM Alain Bensoussan Fellowship Programme, which is supported by the Marie Curie Co-funding of Regional, National and International Programmes (COFUND) of the European Commission. The work is partially supported by Supported by the State Key Program of National Natural Science of China (Grant No. 91324201), the National Natural Science Foundation of China (Grant No. 81271513) of China, the Fundamental Research Funds for the Central Universities of China, the self–determined and innovative research funds of WUT No. 2014-Ia-012, 2013-Ia-040, the Natural Science Foundation No. 2014CFB865 of Hubei Province of China, Scientific Research Foundation for Returned Scholars from Ministry of Education of China (No. 20111j0032), the HOME Program No. 11044 (Help Our Motherland through Elite Intellectual Resources from Overseas) funded by China Association for Science and Technology,The National Soft Science Research Program 2009GXS1D012 of China, the National Science Foundation for Post–doctoral Scientists of China No. 20080431004.

    View full text