Abstract

The application of Shape Memory Alloys with superelastic behavior (SMA–SE) to reduce vibrations in rotor systems has been studied in recent years. However, the current SMA damper device design does not prioritize improving the distribution of applied force due to vibration levels. In this sense, this paper aims to study (theoretical and experimentally) two different types of passive damper devices made of a Ni-Ti SMA–SE to be incorporated in a rotor system in order to reduce mechanical vibrations. Two different physical situations were analyzed in two new designs (A and B) of SMA–SE bending springs. In order to obtain parameters similar to real conditions for the displacement and cycle routines of simulation, experimental tests were made with an unbalanced rotor system first, where the maximum displacement was used for simulations as extreme conditions for this kind of system. The first study was the application of loads and releases, the second is related to applications of preloads and peak-to-peak displacement variations with the aim of verifying real conditions. From the obtained results, it was verified that the damper design of device A reached a better performance (5.63% of specific damping capacity) than that of device B (2.65% of specific damping capacity) during the martensitic transformation, and reaching the full martensitic transformation with less displacement applied for the first setup. For the second setup, the device A reached 12.63% and device B reached 6.44% of the specific damping capacity. Thus, the bending springs of type A are more efficient in reducing mechanical vibration using the damping properties of the SMA-SE.

摘要

近年来，人们研究了具有超弹性行为的形状记忆合金（SMA-SE）在减少转子系统振动方面的应用。然而，当前的 SMA 阻尼器装置设计并没有优先考虑由于振动水平而改善作用力的分布。从这个意义上说，本文旨在研究（理论和实验）两种不同类型的由 Ni-Ti SMA-SE 制成的被动阻尼装置，它们将被结合到转子系统中，以减少机械振动。在 SMA-SE 弯曲弹簧的两种新设计（A 和 B）中分析了两种不同的物理情况。为了获得与模拟的位移和循环程序的真实条件相似的参数，首先对不平衡转子系统进行了实验测试，其中最大位移用于模拟这种系统的极端条件。第一项研究是载荷和释放的应用，第二项研究与预载荷和峰峰值位移变化的应用有关，目的是验证真实情况。从所得结果验证，装置 A 的阻尼器设计在马氏体转变过程中达到了比装置 B 更好的性能（比阻尼能力的 5.63%）（比阻尼能力的 2.65%），并达到了全马氏体在第一次设置中应用较少位移的变换。对于第二个设置，设备 A 达到了 12.63%，设备 B 达到了特定阻尼能力的 6.44%。因此，