汽轮发电机组故障诊断的复合式进化网络

提出了一种复合式神经网络结构，并用于大型汽轮发电机组的故障诊断．该神经网络集成一系列的ＢＰ网络，来完成故障分类任务．每个ＢＰ子网络只有一个输出结点并对应于一种特定的状态．子网络的权值通过基因算法进化确定，从而使训练过程可以实时进行．通过这种方法不仅可以对已知故障进行分类，而且可以对存在的新的故障进行识别．一种基于这种结构的实用诊断系统已经投入使用．

A Hybrid Evolving Neural Network for Diagnosis of Turbogenerator

A hybrid neural network architecture has been proposed for the diagnos is of turbogenerator. A series of BP networks are integrated together to perform the task of fault classification. Each BP sub net has only one output and response to one specific condition. Their weights are evolved by genetic algorithms (GAs), which enables the real time training to be performed. On this approach, classification among known faults can be made and a novel fault, if any, can also be identified, A practical diagnosis system has been developed and put in effect.