Application of Intentional Mistuning to Reduce the Vibration Susceptibility of a Steam Turbine Wheel

Abstract Intentional mistuning is employed for a last stage turbine wheel featuring 51 blades to alleviate both the flutter susceptibility and maximum forced response. Primarily, operations at nominal speed under part-load conditions may cause unfavorable flow conditions facilitating flow separation. As a consequence, the original design intention with identical blades features negative aerodynamic damping ratios with respect to the first bending mode family. In order to prevent any self-excited vibration phenomena, intentional alternate mistuning is utilized to increase the least aerodynamic damping ratio as far as it takes a positive value and hence, to contribute to a stabilization of the rotor. For the purpose of numerically analyzing the vibration behavior, reduced order models are built up, which are based on modal reduction techniques, namely the subset of nominal system modes and the fundamental mistuning model. These types of models conveniently allow for considering both different mistuning distributions in terms of probabilistic analyses and the aeroelastic interaction by means of prescribing aerodynamic damping ratios and aeroelastic natural frequencies of the tuned counterpart or aerodynamic influence coefficients, respectively. A detailed study is presented regarding the correction of frequency mistuning magnitudes in terms of considering the impact of centrifugal stiffening, which plays a significant role in case of long low pressure turbine blades featuring high aspect ratios. Since alternate intentional mistuning cannot be implemented perfectly, every bladed wheel as manufactured will exhibit small but unavoidable structural deviations from the design intention, which are known as random mistuning. To ensure the robustness of the intentional mistuning solution in terms of positive aerodynamic damping ratios at any time, comprehensive probabilistic analyses are conducted with respect to superimposing random structural mistuning at first. Secondly, the impact of varying mistuning magnitude is analyzed. Thirdly, the robustness towards aerodynamic mistuning is investigated by means of small variations of aeroelastic influence coefficients and consequently, the inter blade phase angle dependent aerodynamic damping curves. Moreover, it becomes apparent that alternate intentional mistuning superimposed with both, random structural and aerodynamic mistuning also mitigates the maximum forced response at part-speed conditions.