控制系统响应对液相储罐泄漏场景影响研究

发布时间：2018-04-10 14:27

本文选题：泄漏场景 + HAZOP/LOPA　；参考：《中国石油大学(华东)》2015年硕士论文

【摘要】：储罐泄漏场景中,液位控制系统的响应会加剧泄漏事故的发生,影响保护层独立性,给场景发生频率的计算及风险等级的评估带来较大误差,该现象属于共因失效(Common Cause Failure,CCF)的一种,同时也属于系统性失效(Systematic Failure)的范畴。鉴于共因失效模式的传统分析对象为控制系统元件的随机性失效,对控制系统正常运行情况下的系统性共因失效鲜有研究,使得对此类场景分析的需求愈显迫切,因此,本文将结合共因失效和系统性失效的概念进行液位控制系统对泄漏场景响应的影响分析。首先,本文以液相储罐泄漏场景中的控制系统(BPCS/SIS)为研究对象,对引例中控制系统响应加剧泄漏事故的继续提出假设,带入系统性共因失效概念,并进行归纳;以中国石油大学(华东)管道泄漏及气体扩散测试实验系统作为平台,进行储罐泄漏基本过程控制系统响应实验,通过泄漏强度实验数据总结对比响应前后泄漏强度的变化规律对引例提出的假设进行验证;选取罐体(某大型燃气轮机驱动离心压缩机组旋流器工艺部分)作为案例,利用实验现象及规律,参照国内外相关化工装备HAZOP/LOPA分析体系和方法,针对装置泄漏场景进行LOPA分析,从而对现有控制系统保护层响应情况进行比对,判断此类响应对泄漏事故的正负面影响,提出系统性共因失效模式的产生机制并利用可靠性框图(Reliability Block Diagram,RBD)对其进行研究,同时开展半定量评估分析,得到现有保护层实际的风险降低因子(RRF)以及所需添加的保护层失效概率(Probabilty of Failure on Demand,PFD);类比共因失效(随机性失效)中β因子的意义,建立系统性共因失效的ε因子,对控制系统保护层在设计中存在的系统性共因失效进行评估,并利用马尔科夫(Markov)模型,对存在系统性共因失效的保护层PFD进行分析。最后,研究成果应用于工程实际,针对泄漏场景下控制系统响应后果的不确定性,确定其正负面影响,即在泄漏场景下部分控制系统的响应将导致对比无控制状态的泄漏强度加剧,基于泄漏强度变化规律的实验数据分析,为泄漏场景的LOPA分析提供理论依据及科学参考;结合案例的HAZOP/LOPA分析,利用控制系统泄漏场景影响分析及系统性共因失效分析,对泄漏场景的LOPA分析进行优化,并建立ε因子为衡量控制系统保护层受场景影响的敏感度提供参考。
[Abstract]:In the tank leakage scenario, the response of the liquid level control system will aggravate the leakage accident, affect the independence of the protection layer, and bring great error to the calculation of the occurrence frequency of the scene and the evaluation of the risk level.This phenomenon belongs to the common Cause failure (CCF) and the systematic failure system failure (CCF).Since the traditional analysis object of common cause failure mode is the random failure of control system components, there is little research on systematic common cause failure under normal operation of control system, which makes the demand for this kind of scene analysis more urgent.In this paper, the effects of level control system on the response of leakage scene are analyzed by combining the concepts of common cause failure and systematic failure.Firstly, this paper takes the control system BPCS / SISS in the liquid storage tank leakage scene as the research object, puts forward the hypothesis of the control system response to aggravate the leakage accident in the example, and brings into the concept of systematic common cause failure, and sums up it.Taking the pipeline leakage and gas diffusion test system of China University of Petroleum (East China) as the platform, the response experiment of the basic process control system for the tank leakage is carried out.The hypothesis proposed in this paper is verified by summarizing and contrasting the variation law of leakage intensity before and after response by the experimental data of leakage strength, and the tank (the process of a large gas turbine driven centrifugal compressor hydrocyclone) is selected as an example.According to the experimental phenomenon and law, referring to the HAZOP/LOPA analysis system and method of chemical equipment at home and abroad, the leakage scene of the plant is analyzed by LOPA, and the response of the protective layer of the existing control system is compared.To judge the positive and negative effects of this kind of response on the leakage accident, the mechanism of systematic common cause failure mode is put forward, and the reliability Block diagram is used to study it. At the same time, the semi-quantitative evaluation analysis is carried out.The actual risk reduction factor (RRFF) of the existing protective layer and the failure probability of the protective layer (Probabilty of Failure on demand PFDs) are obtained, and the 蔚 factor of systematic common cause failure is established by analogy with the meaning of 尾 factor in the common cause failure (random failure).The systematic common cause failure in the design of the protective layer of the control system is evaluated, and the PFD of the protective layer with systematic common cause failure is analyzed by using Markov Markov model.Finally, the research results are applied to the engineering practice, aiming at the uncertainty of the response consequences of the control system under the leakage scenario, the positive and negative effects are determined.That is, the response of part of the control system under the leakage scenario will lead to the intensification of the leakage intensity compared with the non-controlled state. The experimental data analysis based on the law of the leakage intensity change provides the theoretical basis and scientific reference for the LOPA analysis of the leakage scene.Combined with the HAZOP/LOPA analysis of the case, the LOPA analysis of the leakage scene is optimized by using the impact analysis of the leakage scene of the control system and the systematic common cause failure analysis.The 蔚 factor is also established to measure the sensitivity of the protection layer of the control system affected by the scene.
【学位授予单位】：中国石油大学(华东)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ086;TP273

【参考文献】