【摘要】 基于一种含外部输入的ARX模型(Env-AR-ARX),提出了一种通过模型匹配的转子故障诊断方法。以航空发动机转子健康状态下的振动数据建立AR模型,模型的准确度是故障诊断的关键。为了提高模型的匹配度,在AR模型的基础上,充分考虑了模型残差和信号时变性的特点,对比讨论了AR模型、AR-ARX模型以及Env-AR-ARX模型,最终确定了最优的Env-AR-ARX模型。由于没有航空发动机转子的故障数据,对本特利转子的故障(碰摩、单裂纹、双裂纹)振动信号建立Env-AR-ARX模型,结合单因素方差分析的方法,确定信号是否存在故障以及故障类型。实验结果表明,该方法能有效应用于转子故障诊断。更多还原

【Abstract】 Based on an ARX model(Env-AR-ARX) including external input, a rotor fault diagnosis method is proposed by model matching. The AR model is built based on the vibration data of healthy aero-engines. The accuracy of the model is the key for the fault diagnosis. To improve the compatibility, the model residual and signal time variation are fully considered based on the AR model. By comparing and discussing AR model, AR-ARX model and Env-AR-ARX model, the Env-AR-ARX model is chosen to be the optimal model. Because of the lack of the fault data of the aero-engine rotors, the faulty vibration signal of the Bentley rotor is used to build the AR model. Using the method of single factor square difference analysis, the existence and type of the faults can be determined. The experiment result shows that this method can be applied effectively for the rotor fault diagnosis.更多还原