This paper investigates numerically and experimentally the rotor drop dynamics when unexpected Active Magnetic Bearings (AMBs) shut down occurs.In such an event, the rotor behaviour drops on two touchdown bearings (TDBs) composed of a ball bearing and a ribbon damper providing stiffness and damping to the overall system. The aim of this paper is to establish and to validate the rotor-drop system model. A first experimental set up is used to identify the dynamic characteristics of the corrugated ribbon damper and to test the Kelvin-Voigt model and the generalized Dahl model. Then, three TDB models are proposed including either the first or the second ribbon damper models. The second experimental set-up is devoted to the rotor drop response of an industrial scale rotor-AMB system equipped with two TDBs. Rotor drop simulations in the time domain are carried out by using the Finite Element method and the three TDB models. Predicted and measured rotor drop responses are compared regarding the displacements as well as transmitted loads and permits evaluating the model efficiency.

The aim of this study was to develop and implement a new control approach dedicated to turbomachinery. The new, fuzzy based controller utilizes inputs expressed in polar coordinates. Its originality is that it manages two significant physical quantities, namely, tangential and radial velocities, associated with steady-state and transient behaviors, respectively. Three controllers are compared for the control of a flexible rotor supported by active magnetic bearings (AMBs): proportional-integral-derivative (PID), single-input and single-output (SISO) fuzzy and the new controller. The assessment was performed using an academic test rig and the results obtained with the new controller show that performances were enhanced with equivalent levels of stability and robustness.

The stability of rotating machinery is a major challenge for the floating production storage and offloading (FPSO) units such as steam turbines or centrifugal compressors. The use of active magnetic bearings (AMBs) in turbomachines enables high operating speeds, active mechatronic system for the diagnostics, and the control and enables downsizing of the whole installation footprint. In case of strong base motions, the rotor can contact its touchdown bearings (TDBs) which are used as emergency and landing bearings. The aim of this study is to assess the stability of a rotating machine supported on AMBs during severe foundation excitation. The combined effect of unbalance forces, base motion excitation, and contact non-linearity on a rotor–AMB system response is analyzed focusing on the capacity of an augmented proportional-integral-derivative controller to maintain the system stable. An academic scale test rig was used for the experimental investigations. The controller was efficient and able to maintain the system stable during and after the application of the excitation, but the dynamic capacity of the AMBs was largely oversized with respect to the studied system. In order to check the capacity of the AMBs, when they are designed as a function of the rotor weight and expected excitation, numerical simulations were carried out (downsized). A finite element (FE) model was developed to model the on-board rotor–AMB system. Predicted and measured responses due to impulse excitation applied on the foundations were compared. The capacity of the controller to maintain the system stability is then discussed.

The aim of this study is to assess the possibility to apply a new control approach dedicated to turbomachinery. The controller is fuzzy based using inputs expressed in polar coordinates. The advantage is that it manages two significant physical quantities, namely tangential and radial velocities that are related to steady state and transient behaviors, respectively. A synchronous filter is associated to the controller in order to enhance the ratio command force/bearing dynamic capacity. The approach was previously applied experimentally with success for the control of an academic test rig. It is adapted here for the control of an industrial compressor whose flexible rotor is supported by active magnetic bearings (AMB). At this stage, only numerical investigations are performed. The controller has to satisfy the standards and the end users requirements. In addition, it should be easy to implement. The behavior of the machine studied is assessed for several configurations of unbalances. A test that corresponds to usual industrial excitations (subsynchronous excitations at nominal speed) is also carried out. Results obtained are satisfactory and give insight into the potential of the approach. In addition, and as the fuzzy controller parameters are independent from the rotor design, the approach is a first step for the standardization of magnetic bearing controller synthesis.

No abstract