论文部分内容阅读
人工神经网络模型是用大量简单的处理单元广泛连接而成的一个非线性动力学网络系统。它以高度的并行分布式处理、联想记忆、自组织及自学习能力和极强的非线形映射能力，为现代故障诊断技术的智能化发展提供了一个全新的方法。介绍了人工神经网络的基本性能和BP网络模型及算法，在此基础上将神经网络和模式识别相结合研究了通过识别抽油机示功图诊断抽油机故障的方法。
Artificial neural network model is a non-linear dynamic network system made up of a large number of simple processing units. Its highly parallel distributed processing, associative memory, self-organization and self-learning capabilities and a strong non-linear mapping capabilities for the intelligent development of modern fault diagnosis technology provides a new way. The basic performance of artificial neural network and BP neural network model and algorithm are introduced. Based on this, the neural network and pattern recognition are combined to study the method of diagnosing the fault of pumping unit by identifying the dynamometer.