摘要
分析了凝汽器工作过程及故障机理,建立了凝汽器典型故障集、征兆集及典型故障特征向量集合。建立了基于支持向量机的故障诊断模型,用实例计算证明其有效性。同时和神经网络方法对比后发现:在小样本情况下,采用支持向量机方法的计算结果比神经网络更优越,推广能力更强,而且效率高于神经网络。本方法针对故障诊断样本少的特点,为建立智能化的凝汽设备状态监控和故障诊断提供了一种新的途径,具有广泛的实用价值。
Condenser's working process and fault mechanism are analyzed, whose typical fault concourses, symptom concourses and typical fault feature vectors are established. A fault diagnosis model is set up based on support vector machine method. The effective of this method has been demonstrated by specific sample calculations. A comparison with a neural network model has shown that the support vector machine method is superior to the neural network method in terms of calculation results, generalization ability and efficiency under the condition of small quantity samples. When a relatively small number of diagnosis samples are involved, the above method may provide a new approach for creating an intelligent system of highly practical value for the condition monitoring and fault diagnosis of condenser.
出处
《热力透平》
2009年第1期57-60,共4页
Thermal Turbine
关键词
汽轮机
凝汽器
支持向量机
神经网络
故障诊断
gas turbine
condenser
support vector machine
neural network
fault diagnosis