Page 493 - Mathematical Techniques of Fractional Order Systems
P. 493

480  Mathematical Techniques of Fractional Order Systems


                          1.5



                           1

                         y

                          0.5



                           0
                           1600    1700     1800    1900    2000
                                             t
            FIGURE 16.4 Presentation of the variable y in system with infinite equilibria (16.10) for two
            different  initial  conditions  ðxð0Þ; yð0Þ; zð0ÞÞ 5 ð0:1; 0:1; 0:1Þ  (blue  solid)  and
            ðxð0Þ; yð0Þ; zð0ÞÞ 5 ð0:1; 0:101; 0:1Þ (red dashed).


               In addition, Kaplan York dimension of the system is calculated by:
                                         L 1 1 L 2
                                D KY 5 2 1      5 2:1158:            ð16:13Þ
                                           jj
                                           L 3
               As has been known, a typical feature of a chaotic system is the sensitive
            dependence on initial conditions. Fig. 16.4 illustrates such a feature of a system
            with infinite equilibria (16.10). We have changed initial conditions slightly
            from ðxð0Þ; yð0Þ; zð0ÞÞ 5 ð0:1; 0:1; 0:1Þ to ðxð0Þ; yð0Þ; zð0ÞÞ 5 ð0:1; 0:101; 0:1Þ.As
            canbeseenin Fig. 16.4, a small change of initial conditions leads to large dif-
            ferences in the system’s state.
               Moreover, we have also verified the chaoticity of a system (16.10) with
            infinite equilibria by using Poincare ´ maps and frequency spectra. Poincare ´
            maps of system (16.10) are presented in Fig. 16.5. Fig. 16.6 shows the fre-
            quency spectra of system (16.10). It is easy to see from Figs. 16.5 and 16.6
            that the system (16.10) with an infinite number of equilibrium points is
            chaotic.
               We have investigated the dynamics of system (16.10) with infinite equi-
            libria by changing the value of the bifurcation parameter b from 0.06 to
            0.12. Figs. 16.7 and 16.8 present the bifurcation diagram and the diagram of
            maximal Lyapunov exponents (MLEs) of the system, respectively. From
            Figs. 16.7 and 16.8, obviously we can see the presence of a period-doubling
            route to chaos when increasing the value of the parameter b. For example,
            system (16.10) with infinite equilibria displays period-1 oscillation for
            b 5 0:07 (Fig. 16.9A), period-2 oscillation for b 5 0:08 (Fig. 16.9B), and
            period-4 oscillation for b 5 0:092 (Fig. 16.9C). Chaotic oscillations can be
            observed for b . 0:934.
   488   489   490   491   492   493   494   495   496   497   498