论文部分内容阅读
人工神经网络模型是在现代生物神经系统研究基础上建立的一种网状结构,是对人脑某些基本特性的一种简单的数学模拟。神经网络以其信息的并行分布式处理、联想记忆、自组织及自学习能力,在机械故障诊断领域显示了极大的应用潜力。本文就神经网络对给定知识的表达、联想、记忆能力及网络结构进行了研究,利用反向误差传播网络对旋转机械中四种典型故障进行了实例分析诊断,取得了令人满意的效果。
Artificial neural network model is a kind of network structure based on the research of modern biological nervous system. It is a simple mathematical simulation of some basic characteristics of human brain. Neural network with its parallel distributed processing of information, associative memory, self-organizing and self-learning capabilities, has shown great potential in the field of mechanical fault diagnosis. In this paper, we study the expression, association, memory and network structure of a given knowledge in neural network. By using the reverse error propagation network, we analyze and diagnose four kinds of typical faults in rotating machinery and obtain satisfactory results.