论文部分内容阅读
基于同步发电机可控硅励磁系统经常发生故障,提出一种基于粗糙集—神经网络(Roughset-NeuralNetwork)相结合的故障诊断方法。以励磁系统中三相桥式可控硅整流回路为核心进行故障诊断研究,对整流回路故障波形的采样数据样本信息进行预处理,通过运用粗糙集理论的知识约简方法形成故障诊断的确定性规则,从而实现故障分类;然后将其结果与故障信息中的输出样本值作为神经网络的输入,实现故障元的定位。通过计算机仿真,结果表明:该方法对三相桥式可控硅整流回路故障诊断简便准确,诊断速度快。
Based on the frequent failures of thyristor excitation system of synchronous generator, a fault diagnosis method based on rough set-neural network (Roughset-NeuralNetwork) is proposed. Taking the three-phase bridge thyristor rectifier in the excitation system as the core, the fault diagnosis is studied. The sample data of the fault waveform in the rectifier loop is preprocessed, and the certainty of the fault diagnosis is formed by the knowledge reduction method using rough set theory Rules, so as to realize fault classification. Then, the output sample values in the result and fault information are used as the input of neural network to locate the faulty element. The result of computer simulation shows that this method is simple and accurate for fault diagnosis of three-phase bridge rectifier thyristor. Its diagnosis speed is fast.