This dissertation is concerned with identification of nonlinear instabilities in rotating multibody systems and subsequent control to eliminate the vibrations. Three nonlinear mechanical systems of this type are investigated and instabilities arising from their inherent nonlinearities are shown to exist for a range of system parameters and conditions. Subsequently, various nonlinear methods of vibration control have been employed to eliminate or suppress the instabilities. Analytical and numerical models have been designed to demonstrate various unstable dynamical behaviour with consistent results. The motion is studied by means of time history, phase space, frequency spectrum, Poincare map, Lyapunov characteristic exponents and Correlation Dimension. Numerical simulations have also shown the effectiveness and robustness of the control techniques over a range of instability conditions for each model. The dynamics of a rotating body with internal energy dissipation is first investigated. Such a model may be considered to be representative of a simplified spinning spacecraft. A comprehensive stability analysis is performed and regions of highly nonlinear behaviour are identified for more rigorous investigation. Numerical simulations using typical satellite parameter values are performed and the system is found to exhibit chaotic motion when a sinusoidally varying torque is applied to the spacecraft for a range of forcing amplitude and frequency. Analysis of this model using Melnikovs method is performed and a sufficient criterion for chaotic instabilities is obtained in terms of system parameters. Evidence is also presented, indicating that the onset of chaotic motion is characterised by period doubling as well as intermittency. Subsequently, Control of chaotic vibrations in this model is achieved using three techniques. The control methods are implemented on the model under instability conditions. The first two control techniques, recursive proportional feedback (RPF) and continuous delayed feedback are recently developed model independent methods for control of chaotic motion in dynamical systems. As such these methods are employed on all three rotating multibody systems in this dissertation. Control of chaotic instability in this model is also achieved using an algorithm derived using Lyapunovs second method. Each technique is outlined and the effectiveness of the three strategies in controlling chaotic motion exhibited by the present system is compared and contrasted. The dynamics of a dual-spin spacecraft with internal energy dissipation in the form of an axial nutational damper is also investigated for non-linear phenomena. The problem involves the study of a body with internal moving parts that is characterised by a coupling of the motions of the damper mass and the angular rotations of the platform and rotor of the spacecraft. Two realistic spacecraft parameter configurations are investigated and each is found to exhibit chaotic motion when a sinusoidally varying torque is applied to the spacecraft rotor for a range of forcing amplitude and frequency. Onset of chaotic motion was shown to be characterised by period doubling and Hopf bifurcations. An investigation of the effects of damping upon the configuration is also performed. Predicted instabilities indicate the range of rotor speeds, perturbation amplitudes and damping coefficients to be avoided in the design of dual-spin spacecraft. Control of chaotic vibrations in this model is also achieved using recursive proportional feedback (RPF) and continuous delayed feedback. Subsequently a more effective model dependent method based on energy considerations is derived and implemented. The effectiveness and robustness of each technique is shown using numerical simulations. Another rotating multibody system that is physically distinct from the previously described models is also investigated for nonlinear instabilities and control. The model is in the form of a driveline which incorporates a commonly used coupling called a Hookes joint. In particular, torsional instabilities due to fluctuating angular velocity ratio across the joint are examined. Linearised equations are used for the prediction of critical speed ranges where parametric instabilities characterised by exponential build up of torsional response amplitudes occur. Predicted instabilities indicate the range of driveshaft speeds to be avoided during the design of a driveline which employs a Hooke's joint. Numerical simulations further demonstrate the existence of parametric, quasi-periodic and chaotic instabilities. Subsequently, suppression of these vibrations is achieved using the previously described model independent techniques. Chaotic vibrations have also been observed in a range of simple mechanical systems such as a periodically kicked rotor, forced pendulum, synchronous rotor, aeroelastic panel flutter and impact print hammer to name but a few. It is thus becoming of increasing importance to engineers to be aware of chaotic phenomena and be able to recognise, quantify and eliminate these undesirable vibrations. The analytical and numerical methods described in this dissertation may be usefully employed by engineers for detecting as well as controlling chaotic vibrations in an extensive range of physical systems.
| Identifer | oai:union.ndltd.org:ADTP/253780 |
| Creators | Meehan, Paul Anthony |
| Source Sets | Australiasian Digital Theses Program |
| Detected Language | English |
Page generated in 0.0016 seconds