%0 Journal Article
%T 动态不确定因果图用于复杂系统故障诊断<br>Fault diagnostics using DUCG incomplex systems
%A 赵越
%A 董春玲
%A 张勤
%J 清华大学学报（自然科学版）
%D 2016
%R 10.16511/j.cnki.qhdxxb.2016.25.012
%X 商用核电站中的操作员在电站正常运行时需要密切监测核反应堆的运行状态。当故障发生时, 对核电站进行迅速、有效的故障诊断和对故障进行正确处理极为重要。该文介绍了动态不确定因果图(dynamic uncertain causality graph, DUCG)理论方法, 并将DUCG方法应用于核电站的故障诊断。以中国广核集团有限公司的宁德核电站1号机组CPR1000为原型建立了8类典型的二回路故障模型, 进行故障诊断验证和故障发展预测。同时, 应用该公司全配置仿真系统对每个故障进行了20次实际测试。验证和测试结果均表明: DUCG能够准确、快速、高效地进行故障诊断。<br>Abstract：The status of nuclear reactors in commercial nuclear power plants needs to be closely monitored to maintain normal operations. When a failure occurs, rapid and effective fault diagnostics and proper handling of failures is extremely important. This paper applies dynamic uncertain causality graph (DUCG) theory to fault diagnostics of nuclear power plants. The method was applied to a model with 8 typical second and loop faults based on the Ningde Nuclear Power Plant Unit 1 CPR1000 of the China Guangdong Nuclear Power Group (CGNPC) to verify the fault diagnostics and initial progression forecasts. Simulations were used to test each fault 20 times. The method and stimulator tests both showed that DUCG can accurately, quickly and efficiently diagnose faults.
%K 动态不确定因果图
%K 复杂系统
%K 故障诊断
%K &lt
%K br&gt
%K DUCG
%K complex system
%K fault diagnosis
%U http://jst.tsinghuajournals.com/CN/Y2016/V56/I5/530