论文部分内容阅读
提出一种利用支持向量机进行直升机旋翼不平衡故障诊断的方法,建立了用于直升机旋翼不平衡故障识别的支持向量机诊断模型,进行了直升机旋翼不平衡故障模拟试验,分别采集了在旋翼配重不平衡、桨距不平衡、后缘调整不平衡和正常状态下的试验台体振动信号,并对其进行了功率谱分析。采用基于支持向量机的诊断模型对旋翼不平衡故障进行了故障分类识别,并与基于径向基神经网络的诊断模型进行了故障识别效果对比。结果表明基于支持向量机的诊断方法在小样本条件下,对旋翼不平衡故障具有良好的识别能力。
A method to diagnose unbalanced rotor helicopter rotor faults using support vector machine is proposed. A support vector machine (SVM) diagnosis model for helicopter rotor unbalance fault identification is established. The helicopter rotor imbalance fault simulation test is carried out. Heavy imbalance, pitch imbalance, trailing edge adjustment unbalance and normal state of the test platform vibration signal, and its power spectrum analysis. Fault diagnosis and classification based on support vector machine (SVM) are used to identify unbalanced rotor faults, and compared with fault diagnosis based on RBF neural network model. The results show that the SVM-based diagnostic method has a good ability of recognizing unbalanced rotor faults under small sample conditions.