论文部分内容阅读
介绍了系统的突变故障和缓变故障,阐述了故障误报率和漏报率等性能指标,指出了故障预测研究的关键问题.对动态系统故障预测的主要方法进行了概述,包括基于神经网络、专家系统和模糊神经网络的故障预测方法,基于灰色系统、粒子滤波器、时间序列分析的故障预测方法.说明了灰色系统的故障预测和基于粒子滤波器故障预测的缺点.分析了人工神经网络预测的优点和缺点.列举了主要故障预测方法的特点和典型应用.对目前故障预测研究现状和存在的问题以及今后研究趋势做了说明.
The catastrophic failures and creeping faults of the system are introduced, the performance indexes such as fault false alarm rate and false negative rate are expounded, and the key problems of fault prediction research are pointed out.The main methods of fault prediction of dynamic system are summarized, including neural network, Expert system and fuzzy neural network, the fault prediction method based on gray system, particle filter and time series analysis, the fault prediction of gray system and the fault prediction based on particle filter are described.The artificial neural network prediction The characteristics and typical applications of the main fault prediction methods are listed, and the current research status and existing problems of the fault prediction and the future research trends are described.