论文部分内容阅读
分析了专家系统ES和前向神经网络MLP对新事件的识别能力的不足 ,探讨了神经网络的数学模型和数学结构 ,建立了MLP和ART1相结合的混合型神经网络模型 ,以实现故障模式的有效识别 .在泵系统故障诊断与监测中的应用表明 ,混合型神经网络能可靠地进行水泵机组故障模式的识别 .
The shortage of expert system ES and the forward neural network MLP for identifying new events is analyzed. The mathematical model and mathematical structure of the neural network are discussed. A hybrid neural network model combining MLP and ART1 is established to realize the failure modes Effective identification.Application in the fault diagnosis and monitoring of pump system shows that the hybrid neural network can reliably identify the fault modes of pump units.