基于粗糙集和神经网络的导弹故障诊断方法

Missile Autopilot Fault Diagnosis Based on Rough Set and Artificial Neural Networks

摘要:

人工神经元网络(ANN)具有本质的非线性特性、并行处理能力以及自组织自学习的能力,但单独使用ANN处理问题时,往往会存在一些缺陷。文章介绍导弹驾驶仪故障智能诊断的一种新方法:首先,利用粗糙集原理约简故障特征属性数据；其次,用带动量项的批处理BP神经网络方法对故障数据进行训练并检验；最后,将故障数据处理后输入神经网络分类器,对故障实施诊断。

Abstract:

Artificial neural networks have the essential nonlinear character, parallel processing ability, and the ability of self organization and self-learning. But when only using ANN to solve a problem, it often has some shortcomings. In this paper, a new intelligence fault diagnosis method on missile autopilot was presented. First, the fault attributions was reduced according to the rough set theory, the BP neural network which absorbed adding momentum ways and batch ways was trained. It made the fault diagnosis more autom atically.