拓本
非线性系统
转子(电动)
振动
控制理论(社会学)
偏转(物理)
直升机旋翼
有限元法
结构工程
运动方程
机械
工程类
计算机科学
物理
机械工程
经典力学
声学
量子力学
人工智能
控制(管理)
作者
K. Prabith,I. R. Praveen Krishna
标识
DOI:10.1016/j.ijmecsci.2021.106861
摘要
The main aim of this paper is to propose a numerical procedure for capturing the nonlinear dynamic characteristics of a two-spool aero-engine rotor system undergoing multi-disk rub-impact. In aero-engines, the possibility for the multi-disk rub-impact is high during the fan blade-out (FBO) event and subsequent windmilling action. It intensifies the nonlinear effects on the rotor vibrations and leads to certain undesired circumstances in the engine. The dual-rotor model consists of multi-stage compressors and single-stage turbines that undergo rubbing whenever their deflection exceeds the clearance. The dynamic model of the dual-rotor system is constructed using the tapered Timoshenko beam elements, rigid disks and rolling contact bearings. A proper model reduction technique based on component mode synthesis coupled with the Craig-Bampton substructuring is utilized to reduce the size of the finite element model. A semi-analytic technique called the approximate time variational method is employed to investigate the steady-state response of the system under multi-disk rub impact. Based on the proposed method, the response characteristics of the model are obtained and are verified with the numerical integration results. Compared to the single-disk rub-impact, the nonlinearities are intensified and significant variations are observed in the response characteristics and stability of the system. Period-5, quasi-periodic, and dry friction backward whirl motions are observed in the response for different values of the system parameters. During quasi-periodic motion, some unknown fractional components such as 0.716ω1, 0.766ω1, 0.916ω1 and 0.964ω1 are appeared in the response. Moreover, the dry friction backward whirl happened with a very large amplitude and it contains a superharmonic frequency component in the response.
科研通智能强力驱动
Strongly Powered by AbleSci AI