【摘要】 为准确评估核电站棒控棒位系统的可靠性，提出一种结合GO法与动态贝叶斯网络(DBN)并融入布尔代数的系统可靠性定量分析方法——GO-DBN拓宽法.该方法首先分析棒控棒位系统的组成与工作原理，采用GO法操作符表示系统部件，建立GO图模型；随后基于布尔代数处理系统中存在的闭环反馈结构；最后将GO法操作符映射至动态贝叶斯网络，进行系统可靠性计算与反向推理，以识别薄弱环节.结果表明，该方法能够更真实地描述系统特性，不仅解决了GO法和DBN在含闭环反馈结构的复杂系统中概率计算的难题，还突破了GO法仅能计算单一时刻系统可靠度的局限，获得了系统可靠度随时间变化的曲线.分析表明，电源柜和反应堆是易导致系统失效的薄弱环节，在预防性维修中应予以优先考虑.更多还原

【Abstract】 To accurately evaluate the reliability of the rod control and position system in nuclear power plants, a quantitative reliability analysis method integrating the GO method and dynamic Bayesian network(DBN) with Boolean algebra, called GO-DBN widening method, is proposed. This method first analyzes the composition and working principle of the rod control and position system, uses GO operators to represent system components, and establishes a GO model. It then addresses the closed-loop feedback structures within the system based on Boolean algebra. Finally, the GO operators are mapped to a dynamic Bayesian network to perform system reliability calculations and backward reasoning for identifying weak links. The results show that this method can more authentically describe system characteristics. It not only resolves the challenge of probability calculation in complex systems with closed-loop feedback structures using the GO method and DBN, but also overcomes the limitation of the GO method, which can only calculate system reliability at a single moment, thereby obtaining a curve of system reliability over time. Analysis indicates that the power cabinet and the reactor are weak links prone to causing system failure, and should be prioritized in preventive maintenance.更多还原