|
Magnetic bearings are very attractive in industrial environments because they are practically frictionless and have very low energy losses. However, they have a tendency to become unstable and require a very sophisticated control system to keep them operational. To be able to avoid catastrophic failures with a high degree of certainty, it is necessary to develop a better understanding of the non-linear phenomena that characterise the behaviour of this class of bearing.
The aim of this project, funded by the Australian Research Council is to gain insight into the effects of non-linearities on the dynamic behaviour of active magnetic bearings(AMBs), and the influence of non-linearities on their performance. Attention is being focussed on stability analysis, bifurcation control, malfunction diagnosis, determination of stable operating conditions, prediction of AMB performance, and on aspects of non-linear dynamic behaviour including bifurcations, coexistence of multiple solutions and amplitude-modulated motions.
Specific aims of the project are to:
1) investigate the effects of geometric coupling and non-linear force relationships on the dynamic behaviour of the AMBs;
2) investigate the effects of the non-linear force relationship incorporating time delays, saturation of magnetic material, saturation of power amplifier, limitation of the control current, and eddy current effect, respectively, on the dynamic behaviour and the performance of the AMBs;
3) investigate the effects of the combination of three or more components of non-linearity on the dynamic behaviour and performance of the AMBs;
4) determine the stability region of the AMBs, the critical values of time delays, and the parameter regimes for normal operation of the AMBs;
5) analyze the non-linear response of rotor-AMB systems under primary, sub-harmonic, and super-harmonic resonance conditions;
6) explore the parameter regions for existence of multiple solutions, investigate local and global bifurcations, and periodically- and chaotically-amplitude modulated responses;
7) develop a non-linear control approach from the viewpoint of bifurcation control to control saddle-node and Hopf bifurcations, thereby eliminating the occurrence of jump phenomena and amplitude modulated responses;
8) investigate the effect of the non-linearities on the dynamic behaviour of self-sensing magnetic bearings, and develop an appropriate model for the precise estimation of the rotor position;
9) investigate the transient and steady state dynamic behaviour of the AMBs during pole failures;
10) investigate the influence of the non-linearities on the controller design;
11) provide a precise model for the design of control system; and
12) provide guidelines for the design of control system after developing a thorough understanding of the non-linear dynamics of the AMBs.
|
|
