【摘要】 针对旋转机械故障的智能诊断问题,提出了结合二阶BP神经网络和模糊网络的集成诊断方法。系统由信号处理子系统和故障诊断子系统两大部分组成。信号子系统负责信号的提取和预处理,诊断子系统则是诊断的核心,不同信号送入不同子神经网络进行诊断,结果通过决策神经网络进行数据融合,做出最后的诊断。该系统具有知识自动获取、识别速度快、鲁棒性及容错能力强等特点。实例证明该系统是有效的。更多还原

【Abstract】 An intelligent fault diagnosis method for rotating machinery is presented,which integrates two-order backpropagation neural network and fuzzy neural network.It consists of signal processing subsystem and diagnostic subsystem.The signals are extracted and preprocessed in the signal pro-cessing subsystem.The diagnostic subsystem is the core,in which different signals are loaded into different sub-neural networks and diagnosed,then the results are sent into decision-making neural network for data fusion and last diagnosis decision.The system obtains knowledge automatically and has the features of fast recognition,better robusticity and error-tolerance.Example test shows its effectiveness.更多还原