基于模型确认的非线性转子支承系统建模方法

A Nonlinear Modeling Method for a Rotor-Support System Based on Model Validation

中文关键词:  转子支承系统  非线性支承  振动测试  非线性参数识别  非线性建模  模态置信准则  航空发动机

英文关键词:rotor-support system  nonlinear support  vibration test  nonlinear parameter identification  nonlinear modeling  modal assurance criterion  aeroengine

中文摘要:

      为了建立能够准确反映转子支承系统真实结构的数学模型，提出基于模型确认辨识转子系统非线性支承刚度参数，并 建立转子支承系统的非线性动力学模型的方法。采用自由状态下的转子模态测试数据对转子模型进行修正与确认，建立准确的 转子模型。在此基础上，对转子支承系统开展低幅值激励的模态测试，辨识其线性支承刚度参数，从而建立转子支承系统的底层 线性模型。对转子支承系统开展不同激励水平下的恒位移振动测试，确定支承非线性刚度的函数形式，并进一步识别支承非线性 刚度参数，最终建立转子支承系统的非线性动力学模型。设计加工了考虑非线性支承因素的转子系统，采用该方法建立的转子支 承系统底层线性模型，与实际结构的模态频差在5%以内，模态置信准则（MAC）值在0.89以上，具有良好的相关性；进一步建立转 子支承系统的非线性动力学模型，将模型预测的响应与转动试验数据进行对比，结果表明：所提出的方法能够基本准确地反映真 实结构的非线性动力学特性。

英文摘要:

      To establish a mathematical model that accurately reflects the real structure of the rotor-support system, a method based on model validation was proposed to identify the nonlinear support stiffness parameter and establish the nonlinear dynamic model for a rotor-support system. An accurate rotor model was established by using free-state rotor modal test data to modify and validate the rotor model. On this basis, low amplitude excitation modal tests were carried out on the rotor-support system to identify the linear support stiff? ness parameters, thereby the underlying linear model for the rotor-support system was established. The constant displacement vibration tests under different excitation levels were carried out on the rotor-support system to determine the functional form of the nonlinear support stiffness and further identify the nonlinear support stiffness parameters. Finally, the nonlinear dynamic model for the rotor-support system was established. A rotor system considering nonlinear support factors was designed and fabricated, and vibration tests were conducted, the model-predicted responses were compared with test data, the results show that for the established linear model, the modal frequency differ? ences are within 5%, and the Modal Assurance Criterion (MAC) values are above 0.89, showing good correlations; the established nonlinear model can accurately reflect the nonlinear dynamic characteristics of the real structure.