基于GO-FLOW方法反应堆净化系统动态可靠性分析

传统上的系统风险分析实质上是一种静态分析方法(如事件树/故障树方法),在描述影响系统状态变化的诸多因素方面存在诸多困难,如时间、过程变量、硬件状态、历史场景和人因等。研究将GO-FLOW用于动态特性显著的反应堆净化系统可靠性分析,改进备用单元的GO-FLOW动态模型,构建的系统GO-FLOW模型图有效模拟了初因事件下系统结构和配置的改变以及操作员的状态和行为。实例分析表明,在初因扰动下反应堆净化系统的失效概率变化显著,操作员的干预有效降低了因硬件失效所导致的系统风险;GO-FLOW是分析包含人因的动态系统的有效而实用的方法。

Dynamic Reliability Analysis of the Nuclear Reactor Purification System Based on GO-FLOW Methodology

The conventional methods used for system risk analysis are static in nature,such as event tree or fault tree method.To describe the system state variation,we must take multiple factors into account,which include time,procedure variables,hardware,historical events and human factor.For this purpose,GO-FLOW was applied to the reliability analysis of the nuclear reactor purification system with obvious dynamic characteristics.GO-FLOW dynamic model of standby unit was improved,and established system GO-FLOW chart has effectively modeled the changing process of system＇s structure and configuration,as well as the operator＇s quality state and behavior under initial event.The analysis results show that the probability of the purification system failure changes significantly and the intervention by the operator can effectively mitigate the system risk due to hardware failure.It is also verified that the GO-FLOW methodology is an effective and useful analysis tool for dynamic system.