Nonlinear dynamics of a classical rotating pendulum system with multiple excitations
Ning Han(韩宁)† and Pei-Pei Lu(鲁佩佩)
Key Laboratory of Machine Learning and Computational Intelligence, College of Mathematics and Information Science, Hebei University, Baoding 071002, China
We report an attempt to reveal the nonlinear dynamic behavior of a classical rotating pendulum system subjected to combined excitations of constant force and periodic excitation. The unperturbed system characterized by strong irrational nonlinearity bears significant similarities to the coupling of a simple pendulum and a smooth and discontinuous (SD) oscillator, especially the phase trajectory with coexistence of Duffing-type and pendulum-type homoclinic orbits. In order to learn the effect of constant force on this pendulum system, all types of phase portraits are displayed by means of the Hamiltonian function with large constant excitation especially the transitions of complex singular closed orbits. Under sufficiently small perturbations of the viscous damping and constant excitation, the Melnikov method is used to analyze the global structure of the phase space and the feature of trajectories. It is shown, both theoretically and numerically, that this system undergoes a homoclinic bifurcation and then bifurcates a unique attracting rotating limit cycle. Finally, the estimation of the chaotic threshold of the rotating pendulum system with multiple excitations is calculated and the predicted periodic and chaotic motions can be shown by applying numerical simulations.
Received: 06 April 2020
Revised: 09 June 2020
Accepted manuscript online: 18 June 2020
Fund: the National Natural Science Foundation of China (Grant Nos. 11702078 and 11771115), the Natural Science Foundation of Hebei Province, China (Grant No. A2018201227) and the High-Level Talent Introduction Project of Hebei University, China (Grant No. 801260201111).
Ning Han(韩宁) and Pei-Pei Lu(鲁佩佩) Nonlinear dynamics of a classical rotating pendulum system with multiple excitations 2020 Chin. Phys. B 29 110502
Fig. 1.
(a) The physical model of the classical rotating pendulum with multiple excitations, (b) the simple pendulum with rotation feature, (c) the mass-spring oscillator with bi-stable characteristics (SD oscillator).
Fig. 2.
When f0 = 0 and ξ = 0: (a) the equilibrium bifurcation set (λ,q), (b) the nonlinear restoring force for (λ,q) ∈ I, (c) the nonlinear restoring force for , (d)–(f) three types of the nonlinear restoring forces for (λ,q) ∈ II, where .
Fig. 3.
(a) Phase portrait with the pendulum-type homoclinic orbits for (λ,q)=(2,3) ∈ I, (b) phase portrait with the coupling of the pendulum-type and Duffing-type homoclinic orbits for (λ,q) = (1.2,3) ∈ II.
Fig. 4.
(a) Cylindrical phase portrait with the pendulum-type homoclinic orbits for (λ,q) = (2,3) ∈ I, (b) cylindrical phase portrait with the coupling of the pendulum-type and Duffing-type homoclinic orbits for (λ,q) = (1.2,3) ∈ II.
Fig. 5.
Dynamic behavior of system (3) for (λ,q) ∈ I: (a) the equilibrium bifurcation diagram for f0 versus x, (b)–(e) the corresponding phase portraits and the nonlinear restoring forces marked by dashed lines.
Fig. 6.
Dynamic behavior of system (3) for (λ,q) ∈ II1: (a) the equilibrium bifurcation diagram for f0 versus x, (b)–(i) the corresponding phase portraits and the nonlinear restoring forces marked by dashed lines.
Fig. 7.
Dynamic behavior of system (3) for : (a) the equilibrium bifurcation diagram for f0 versus x, (b)–(f) the corresponding phase portraits and the nonlinear restoring forces marked by dashed lines.
Fig. 8.
Dynamic behavior of system (3) for (λ,q) ∈ II2: (a) the equilibrium bifurcation diagram for f0 versus x, (b)–(i) the corresponding phase portraits and the nonlinear restoring forces marked by dashed lines.
Fig. 9.
Dynamic behaviors of the pendulum system with viscous damping and constant excitation for (λ,q) ∈ I: (a) none of limit cycle surrounding the equilibrium for 2 π f0 – a1ξ < 0, (b) the bifurcation of the upper homoclinic orbit for 2πf0 – a1ξ = 0, (c) a rotating limit cycle for 2πf0 – a1ξ > 0.
Fig. 10.
Dynamic behaviors of the pendulum system with viscous damping and constant excitation for (λ,q) ∈ II: (a) none of limit cycle surrounding the equilibrium for 2πf0 – a1ξ < 0, (b) the bifurcation of upper homoclinic orbit for 2πf0 – a1ξ = 0, (c) a rotating limit cycle for 2πf0 – a1ξ > 0.
Fig. 11.
Cylindrical phase portraits of the pendulum system with viscous damping and constant excitation for q = 0, ξ = 0.2 and a1 = 8: (a) no limit cycle for f0 = 0.0001, (b) the bifurcation of the upper homoclinic orbit for f0 ≈ 0.2527, (c) a rotating limit cycle for f0 = 0.3.
Fig. 12.
The chaotic boundaries detected by the Melnikov analysis for the Duffing-type and pendulum-type homoclinic orbits at q = 3 and λ = 1.5: (a) σ = 0.5 (, , ), (b) σ = 1 (, , ), (c) σ = 2.
Fig. 13.
When q = 3, λ = 1.5, ξ = 0.06, f0 = 0.03 and ω = 1.8: (a) bifurcation diagram for f1 versus x with the threshold values (solid line), (dashed line) and (dotted line), (b) the corresponding largest Lyapunov exponents under the initial condition (1,0); (c) the phase portraits of a pair of period-1 solutions for f1 = 0.1 and their time histories (d); (e) the phase portrait of period-3 solution for f1 = 0.65 and the corresponding time history (f); (g) the phase portrait of period-13 solution for f1 = 1.5 and the corresponding time history (h); (i)–(k) three chaotic attractors for f1 = 0.92, 1.7, 1.9, respectively.
Fig. 14.
When q = 3, λ = 1.5, ξ = 0.05, f0 = 0.05 and ω = 0.5: (a) bifurcation diagram for f1 versus x with the threshold values (solid line), (dashed line) and (dotted line); (b) the local enlargements for f1 ∈ [1.5,1.8]; (c)–(e) three types of chaotic motions for f1 = 0.8, 1.55, 1.8 and the corresponding time histories (f)–(h); (i)–(k) three periodic motions for f1 = 1.575, 1.616, 1.78, respectively.
Fig. 15.
(a) A two-dimensional parameter space plot for q = 3, λ = 1.5, f0 = 0.1, ξ = 0.2 in the range of f1 ∈ [0,2] and ω ∈ [1,4] and theoretical boundary derived by the Melnikov analysis, (b) the bifurcation diagram of f1 with the threshold f1 = 0.13 and (c) the corresponding Lyapunov exponents for ξ = 1.515 (the gray region corresponds to the chaotic region, the white region represents non-chaotic region and the black curve is the theoretical boundary of chaos).
Fig. 16.
Cylindrical phase portraits of oscillations starting initial condition (x,y) = (1,0): (a) SD-type chaotic motion, (b) Duffing-type chaotic motion; chaotic cylindrical phase portraits of the coupling of oscillations and rotations (c) and (d).
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.