基于多分类SVM的航空逆变器故障诊断

航空逆变器的可靠性对飞机供电系统的安全性和稳定性尤为重要，但当前对于航空供电器的故障诊断的研究较少，无法为航空逆变器提供有效保障。因此，提出基于多分类支持向量机的故障诊断方法，对航空逆变器的多种故障模式进行诊断。针对故障特征耦合性高的问题，采用主成分分析方法提取故障特征，获取低维度的关键特征。由于逆变器具有多种故障模式，且具有非线性的特点，故采用多分类支持向量机算法进行故障诊断。该算法具有极强的分类能力，是处理小样本、非线性问题的有力工具。实验结果表明，该算法模型可对航空逆变器多种工况条件下的15种故障模式进行有效诊断，并且方法诊断速度快，提高了航空供电系统的安全性。

Fault Diagnosis of Aviation Inverter Based on Multi-Classification SVM

The reliability of aviation inverter is particularly important for the security and stability of the aircraft power supply system.However,there is little research on the fault diagnosis of aviation power supplies,which cannot provide effective guarantee for aviation inverter.Therefore,a fault diagnosis method based on multi-classification support vector machine is proposed to diagnose various fault modes of aviation inverter.In order to solve the problem of high coupling of fault features,the principal component analysis method is used to extract fault features and obtain key features of low dimension.Because the inverter has a variety of fault modes and nonlinear characteristics,the multi-classification support vector machine algorithm is adopted for fault diagnosis.The algorithm has strong classification ability,which is a powerful tool for dealing with small samples and nonlinear problems.The experimental results show that the algorithm model can effectively diagnose 15 fault modes of the aviation inverter under various working conditions,and the diagnosis speed is fast,which improves the safety of aviation power supply system.