基于灰熵优化加权灰色关联度模型的核动力装置故障诊断技术  

Fault Diagnosis Technology of Nuclear Power Plant Based on Weighted Degree of Grey Incidence of Optimized Entropy

作　　者：孔衍[1] 任鑫[2] 王川[2] 李振杰[1] 

机构地区：[1]海军蚌埠士官学校,安徽蚌埠233012 [2]海军医学研究所,上海200433

出　　处：《原子能科学技术》2012年第B09期361-365,共5页Atomic Energy Science and Technology

基　　金：海军医学研究所基金资助项目(12HY21)

摘　　要：核动力装置作为一种复杂的灰色系统,故障和征兆间并无确定的对应关系,这使故障诊断变得十分困难。针对该问题,本文提出了基于灰熵优化加权灰色关联度的模型,并建立了故障诊断系统。为验证系统有效性,以核动力装置冷凝器典型故障为例进行了仿真研究。结果表明:该系统诊断结论准确可靠、计算速度较快,可满足实时诊断要求,且具有稳定性好、故障分辨率高等特点。Nuclear power plants（NPPs） are very complex grey system,in which faults and signs have not certain corresponding connection,so it＇s hard to diagnose the faults.A model based on weighted degree of grey incidence of optimized entropy was proposed according to the problem.To validate the system,some simulation experiments about the typical faults of condenser of NPPs were conducted.The results show that the system＇s conclusion is right,and the system＇s velocity is fast which can satisfy diagnosis in real time,and with the distinctive features such as good stability,high resolution rate and so on.

关 键 词：加权灰色关联度 核动力装置 故障诊断 冷凝器 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]