核电厂数字化仪控系统动态概率安全评价方法研究
发布时间：2018-09-03 11:15
【摘要】：动态概率安全评价(PSA)方法是指在系统可靠建模时通过考虑系统演化的时间因素来解释触发或随机逻辑事件之间耦合的方法,它能弥补传统PSA方法的不足,更有效地验证数字化仪控系统的可靠性及安全性。我们对美国核管会报告NUREG/CR-6901指出的最适合分析数字化仪控系统可靠性的两种方法——马尔可夫/区间映射技术(Markov/CCMT)和动态流图机制(DFM)进行研究。 本文先假设系统在满功率工况运行时发生主给水管道小破口事件,应用Markov/CCMT机制对AP1000主给水简化系统进行PSA分析。在分析过程中,选取主给水泵和主给水调节阀作为失效设备,首先得到系统各个状态之间的转移概率,其次根据系统状态考虑所有可能的失效路径,并通过3KeyMaster仿真平台建立的系统模型对这些路径进行模拟,之后基于模拟数据计算得到区间转移概率,进而得到顶事件——蒸汽发生高液位和低液位的概率。应用Markov/CCMT机制可以避免丢失系统某些重要的失效情况,还可解决顶事件的竞争问题,并且得到系统可能发生事故的确切时间。 本文还考虑CPR1000主给水系统在满功率工况下由于设备失效对蒸汽发生器液位所造成的影响,应用DFM机制对其进行PSA分析。首先选取系统关键的物理变量构建DFM模型,同时基于CPR1000核电站全范围仿真机平台生成的数据建立决策表及划分节点状态,分别进行演绎分析和归纳分析。通过演绎分析,得到顶事件——蒸汽发生器发生高高液位的质蕴含项。分析质蕴含项可解释造成顶事件的所有可能因素,同时结合设备的不同失效模式,可指导操作人员在事故发生前采取适当措施引导相关参数向有利的方向演变,避免事故发生。归纳分析过程类似于进行故障模式及影响分析(FMEA),可得到不同初始条件下对系统产生的影响,根据结果改进系统设计以提高安全性能。 通过本文内容的研究,我们掌握了Markov/CCMT和DFM机制的理论和方法应用,可为开发自主知识产权的动态PSA分析软件提供理论基础。
[Abstract]:Dynamic probabilistic security evaluation (PSA) is a method to explain the coupling between trigger and random logical events by considering the time factor of system evolution in reliable system modeling. It can make up for the shortcomings of traditional PSA method. More effectively verify the reliability and security of digital instrument control system. We study two methods of analyzing the reliability of digital instrument and control system-Markov / interval mapping (Markov/CCMT) and dynamic flow graph mechanism (DFM), which are most suitable for the reliability analysis of digital instrument and control system as pointed out by the NUREG/CR-6901 report of the American Nuclear Regulatory Commission. This paper first assumes that the small break event of the main water supply pipeline occurs when the system is running under the full power condition, and applies the Markov/CCMT mechanism to the PSA analysis of the AP1000 main water supply simplified system. In the process of analysis, the main feed water pump and the main water supply regulating valve are selected as the failure equipment, first, the transfer probability between the various states of the system is obtained, and then all possible failure paths are considered according to the system state. These paths are simulated by the system model established by the 3KeyMaster simulation platform, and then the interval transfer probability is calculated based on the simulation data, and the probability of high and low liquid level occurring in the top event-steam is obtained. The application of the Markov/CCMT mechanism can avoid the loss of some important failures of the system, solve the competition problem of the top event, and obtain the exact time when the system may occur an accident. In this paper, considering the influence of the equipment failure on the liquid level of the steam generator in the CPR1000 main feed water system under the full power condition, the PSA analysis is carried out by using the DFM mechanism. Firstly, the key physical variables of the system are selected to construct the DFM model. At the same time, based on the data generated by the CPR1000 nuclear power plant full-scope simulator platform, the decision table and node state are established, and the deductive analysis and inductive analysis are carried out respectively. Through deductive analysis, the content term of the top event-the high liquid level of the steam generator is obtained. The analysis of the qualitative implication term can explain all the possible factors causing the top event, and combined with the different failure modes of the equipment, it can guide the operator to take appropriate measures to guide the relevant parameters to change to a favorable direction before the accident occurs, and to avoid the accident. The process of induction and analysis is similar to that of fault mode and impact analysis (FMEA), can obtain the influence on the system under different initial conditions. According to the results, the system design is improved to improve the security performance. Through the research of this paper, we have mastered the theory and method application of Markov/CCMT and DFM mechanism, which can provide the theoretical basis for developing dynamic PSA analysis software of independent intellectual property rights.
【学位授予单位】：厦门大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM623
本文编号：2219820
[Abstract]:Dynamic probabilistic security evaluation (PSA) is a method to explain the coupling between trigger and random logical events by considering the time factor of system evolution in reliable system modeling. It can make up for the shortcomings of traditional PSA method. More effectively verify the reliability and security of digital instrument control system. We study two methods of analyzing the reliability of digital instrument and control system-Markov / interval mapping (Markov/CCMT) and dynamic flow graph mechanism (DFM), which are most suitable for the reliability analysis of digital instrument and control system as pointed out by the NUREG/CR-6901 report of the American Nuclear Regulatory Commission. This paper first assumes that the small break event of the main water supply pipeline occurs when the system is running under the full power condition, and applies the Markov/CCMT mechanism to the PSA analysis of the AP1000 main water supply simplified system. In the process of analysis, the main feed water pump and the main water supply regulating valve are selected as the failure equipment, first, the transfer probability between the various states of the system is obtained, and then all possible failure paths are considered according to the system state. These paths are simulated by the system model established by the 3KeyMaster simulation platform, and then the interval transfer probability is calculated based on the simulation data, and the probability of high and low liquid level occurring in the top event-steam is obtained. The application of the Markov/CCMT mechanism can avoid the loss of some important failures of the system, solve the competition problem of the top event, and obtain the exact time when the system may occur an accident. In this paper, considering the influence of the equipment failure on the liquid level of the steam generator in the CPR1000 main feed water system under the full power condition, the PSA analysis is carried out by using the DFM mechanism. Firstly, the key physical variables of the system are selected to construct the DFM model. At the same time, based on the data generated by the CPR1000 nuclear power plant full-scope simulator platform, the decision table and node state are established, and the deductive analysis and inductive analysis are carried out respectively. Through deductive analysis, the content term of the top event-the high liquid level of the steam generator is obtained. The analysis of the qualitative implication term can explain all the possible factors causing the top event, and combined with the different failure modes of the equipment, it can guide the operator to take appropriate measures to guide the relevant parameters to change to a favorable direction before the accident occurs, and to avoid the accident. The process of induction and analysis is similar to that of fault mode and impact analysis (FMEA), can obtain the influence on the system under different initial conditions. According to the results, the system design is improved to improve the security performance. Through the research of this paper, we have mastered the theory and method application of Markov/CCMT and DFM mechanism, which can provide the theoretical basis for developing dynamic PSA analysis software of independent intellectual property rights.
【学位授予单位】：厦门大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM623
【参考文献】
相关期刊论文 前1条
1 余少杰;赵军;童节娟;;DFM在人员可靠性分析中的应用[J];核动力工程;2011年04期
，本文编号：2219820
本文链接：https://www.wllwen.com/falvlunwen/zhishichanquanfa/2219820.html
