航空发动机传感器故障与部件故障诊断技术

结合局部学习思想与集成学习技术,提出了一种基于支持向量机-极端学习机-卡尔曼滤波器(SVM-ELM-KF,Support Vector Machine-Extreme Learning Machine-Kalman Filter)的航空发动机传感器故障与突发性部件故障诊断的方法.将改进的迭代约简最小二乘支持向量回归机训练技术推广到分类机中,用于区分传感器故障与部件故障,使得该分类机具有一定的稀疏性.对于传感器故障,利用ELM分类机对故障进行定位.对于部件故障,利用改进的卡尔曼滤波器对发动机各部件的健康参数进行估计,从而对部件故障进行定位.仿真结果表明,提出的故障诊断方法能够准确地区分传感器故障和部件故障,实现故障的有效定位,验证了方法的可行性.

Fault diagnosis for sensors and components of aero-engine

According to local learning and ensemble learning technologies, a method for sensors fault and abrupt components fault diagnosis of aero-engine was proposed based on support vector machine-extreme learning machine-Kalman filter (SVM-ELM-KF). The training approach of improved recursive reduced-least squares support vector regression (IRR-LSSVR) was extended to classification machine to distinguish sensor faults and component faults. The training method makes the classification machine have better sparsity. Considering sensors fault, the ELM was used for fault location. For components fault, the improved KF was adopted for health parameters estimation and fault location. Simulation results show that the proposed method for fault diagnosis can distinguish sensor faults and abrupt component faults accurately, and locate the faults effectively. That is, the proposed method is valid.