本文针对大型水轮机组状态监测与故障诊断系统 ,对采用神经网络进行故障模式识别的方法进行研究 ,为克服单一神经网络模型诊断方法的局限性 ,对现有的网络模型进行了分析和比较 ,提出并探讨了多种神经网络技术应用于故障诊断的诊断方法。该方法综合了多层感知模型和自适应谐振网络各自的优势 ,建立了适用于水轮机组故障诊断的混合网络模型 ,并已有效地应用于水轮机组状态监测与故障诊断系统 In order to overcome the limitation of the single neural network model diagnosis method, this paper analyzes and compares the existing network models, and proposes a method of fault pattern recognition based on neural network for condition monitoring and fault diagnosis system of large turbine. And discusses a variety of neural network technology used in fault diagnosis diagnosis. The method combines the advantages of multi-layer perceptual model and adaptive resonance network, establishes a hybrid network model suitable for fault diagnosis of hydro-turbine and has been effectively applied to the condition monitoring and fault diagnosis system of hydro-turbine