针对舵机故障样本数量不足、诊断知识不完备的实际情况,提出一种基于支持向量机(support vector machine,SVM)的故障诊断方法.根据航舵故障输入输出映射非线性的特点,分析SVM的分类机理,对基于SVM 的故障诊断步骤进行介绍,解决了小样本模式的分类问题,并通过仿真对该方法的有效性进行验证.仿真结果表明:该方法对舵机故障分类准确性可达92%.%Aiming at lack of fault samples and diagnosis knowledge in nautical steer, introduce a method based on support vector machine (SVM). According to nautical steer nonlinear input-output mapping feature, analyze SVM classification mechanism, introduce the fault diagnosis steps based on SVM, and solve the classification problems of small sample mode. Then, use simulation to verify the validity. The simulation result shows that the method fault classification correctness can reach 92%.
展开▼