基于SDG-ES的核动力装置故障诊断技术研究

发布时间：2018-03-24 13:06

本文选题：核动力装置　切入点：故障诊断　出处：《哈尔滨工程大学》2014年硕士论文

【摘要】：由于核动力装置运行时有高剂量的放射性,发生事故后会对工作人员和环境造成巨大的威胁,因此其运行安全性一直备受世人关注。2011年3月发生的日本福岛核事故再次滞缓了核工业的发展。保障核电厂安全运行一直是世界各国共同的目标,故障诊断技术是提高系统安全性的一种有效手段,也是核安全领域研究的热点。核动力装置系统庞大、结构复杂,难以通过建立精确的数学模型来诊断故障。符号有向图(Signed Directed Graph, SDG)作为一种定性的故障诊断方法,能表达复杂的因果关系,适用于核动力装置这样复杂系统的故障诊断。本文在总结前人关于SDG故障诊断方法的基础上,深入研究了传统SDG诊断方法的特点,提出了建立核动力装置SDG模型的方法,并以压水堆一回路系统为研究对象建立了模型。通过引入故障显现时间有效地解决了 SDG诊断过程中模型出现闭环无法诊断的问题;为了提高诊断速度和分辨率,采用SDG与专家系统相结合的方法进行诊断,给出了根据SDG模型提取专家系统规则的方法;对多源故障诊断时将SDG模型图直接作为专家系统的知识表达,克服了传统专家系统获取知识的困难。在上述理论研究基础上,用C#3.5语言结合SQL Server 2005数据库技术开发了基于SDG-ES的核动力故障诊断专家系统。测试结果表明,该系统具有人机界面友好、诊断快速、易于操作等优点。
[Abstract]:As a result of high doses of radioactivity in the operation of nuclear power units, accidents can pose a great threat to staff and the environment, As a result, the safety of nuclear power plants has been attracting worldwide attention. The Fukushima nuclear accident in Japan in March 2011 has once again slowed down the development of the nuclear industry. Ensuring the safe operation of nuclear power plants has always been the common goal of all countries in the world. Fault diagnosis technology is an effective means to improve the safety of the system, and it is also a hot topic in the field of nuclear safety. It is difficult to diagnose faults by establishing accurate mathematical models. As a qualitative fault diagnosis method, symbolic directed graph (SDGs) can express complex causality. It is suitable for the fault diagnosis of complex systems such as nuclear power plants. Based on summarizing the previous methods of SDG fault diagnosis, this paper deeply studies the characteristics of traditional SDG diagnosis methods, and puts forward a method to establish the SDG model of nuclear power plants. In order to improve the diagnosis speed and resolution, the model of PWR primary circuit system is established. The problem of closed loop failure in the process of SDG diagnosis is effectively solved by introducing the time of fault appearance, and in order to improve the speed and resolution of the diagnosis, the problem of closed loop failure in the diagnosis process of PWR is effectively solved. The method of combining SDG with expert system is used to diagnose, and the method of extracting expert system rules according to SDG model is given, and the SDG model diagram is taken as the knowledge expression of expert system directly in multi-source fault diagnosis. This paper overcomes the difficulty of traditional expert system to acquire knowledge. Based on the above theoretical research, a nuclear power fault diagnosis expert system based on SDG-ES is developed by using C#3.5 language and SQL Server 2005 database technology. The test results show that, The system has the advantages of friendly man-machine interface, fast diagnosis and easy operation.
【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP277;TM623;TP311.52

【参考文献】