Modelling, Monitoring and Vibration Control for High Performance Turbomachinery
In order to support the design process of High Performance Turbomachinery with respect to a safe and reliable operation, lateral as well as torsional vibrations are calculated by means of powerful numerical tools. This includes the determination of eigenvalues (natural frequencies, modes and stability) and the vibration amplitudes of the unbalance response for lateral vibrations as well as the shaft stresses and torques due to electrical faults for torsional vibrations. It is well known that the dynamic behavior of a shaft train not only depends on the external excitations and the dynamic characteristics of the rotating parts, but also on some important interaction effects, e. g. Rotor Structure Interaction (Foundation, Casing), Rotor Fluid Interaction (Bearings and Seals), Rotor Blade Interaction (e.g. Last Stage Turbine Blades), Thermomechanical Interaction (Spiral Vibrations due to Rubbing) and Electromechanical Interaction (Electrical Air Gap torques). As a consequence multi-physical methods with tools from different disciplines are necessary. Some of the methods and tools, which are used for the described tasks are: Finite Elements (FE) for Structural Dynamics, Computational Fluid Dynamics (CFD), Electro Dynamics and Heat transfer. Furthermore Reduction methods, Modal Analysis, Identification and Sensitivity Analysis are also of importance. Modelling for Simulation is therefore an important first task for the mechanical engineer. The keynote lecture will present some of the scientific methods and tools, which have been developed and used for typical interaction effects, influencing the dynamic behavior.
When a Turbomachine goes into service, the user expects a stable long term operation with acceptable vibration values. This important second task can be achieved by means of known Monitoring Systems, consisting of sensors and suited signal processing units. Vibrations can be measured and evaluated by features, where Vibration Standards (e.g. ISO or API) can be used for this evaluation. In the keynote some extended Monitoring possibilities will be presented, which may improve the diagnosis for failures. If for example not only vibration responses (system output) are measured, but also excitation forces (system input) via active elements (AMB’s or Piezos), Frequency Response Functions (FRF) can be determined. They tell much more about the dynamic behavior of the rotor system. By comparison of measured and simulated vibration values possible parameter changes can be identified (Diagnosis). In this way the developed model for the dynamic behavior of the Turbomachine can run in parallel to the real system as a digital twin (Model Based Monitoring).
If vibrations in operation do not fulfil the expected behavior, Vibration Control is needed, in order to bring the vibration values back to acceptable values. Classical solutions of vibration control can be subdivided into: reduction of excitation , tuning of system parameters, damping enhancement, vibration absorption and isolation. These solutions can be applied either passive, semi-active or active. A good dynamic model is again very helpful to find optimal solutions. In the keynote some examples will be presented, controlling the vibrations of Turbomachines by using passive solutions (Balancing, Squeeze Film damper) and active solutions (AMB’s and Piezo-Actuators) as well.