基于知识库的核电厂压力变送器可靠性评估
Reliability Assessment of Nuclear Power Plant Pressure Transmitters Based on Knowledge Base

针对核电厂压力变送器可靠性评估中传统方法(如鉴定试验、电路仿真)难以量化元器件的随机失效对压力变送器可靠性影响的问题，提出一种基于知识库的模型驱动可靠性评估方法。采用FIGARO建模语言构建描述压力变送器电路元器件失效模式与后果的知识库；基于知识库在Model Builder平台中构建图形化描述模型并自动生成故障树；结合FIDES可靠性模型计算元器件失效率，利用XFTA和YAMS软件分别开展故障树定量分析与蒙特卡洛仿真。结果表明，严重事故后360小时时，压力变送器不可用度为0.00415858 (XFTA)与0.00405161 (YAMS)，两者相对误差为2.58%，同时最小割集分析表明运算放大器为电路关键元器件。该方法验证了模型驱动方法在复杂仪表系统可靠性评估中的有效性，可提高核电厂关键仪表可靠性评价的规范性和效率。
To address the challenge in nuclear power plant pressure transmitter reliability assessment where traditional methods (e.g., qualification testing, circuit simulation) struggle to quantify the impact of random component failures on system reliability, this study proposes a knowledge base-driven model-based reliability assessment approach. By employing the FIGARO modeling language, a knowledge base describing failure modes and consequences of pressure transmitter circuit components is constructed. Based on this knowledge base, a graphical descriptive model is developed in the Model Builder platform, from which a fault tree is automatically generated. Component failure rates are calculated using the FIDES reliability model, followed by quantitative fault tree analysis and Monte Carlo simulation through XFTA and YAMS software, respectively. Results demonstrate that the system unavailability at 360 hours post-severe accident is 0.00415858 (XFTA) and 0.00405161 (YAMS), with a relative error of 2.58%. Minimal cut set analysis further identifies operational amplifiers as critical components in the circuit. This method validates the effectiveness of the model-driven approach for reliability assessment of complex instrumentation systems, significantly enhancing the standardization and efficiency of reliability evaluation for critical nuclear power plant instruments.