针对故障假设分析能有效辨识各种潜在危险和可能事故的优势,保护层分析能对事故后果进行定量分析并能核实已有安全措施有效性大小的特点,提出将两者集成,有利于取长补短,更好地实现风险评价和风险控制的目的。介绍了故障假设分析和保护层分析基本方法,阐述了两种方法集成的可行性、集成衔接关系、集成方法评价步骤以及集成方法的作用,并通过实例说明了集成方法的功能。结果表明:故障假设分析和保护层分析方法的集成,不仅能有效识别潜在风险事故,而且能对事故后果及已有安全措施的可靠性进行量化评估,以确定风险的可接受性和安全措施的有效性,进而达到确认是否需要增加新的安全措施的目的。
The paper intends to present a new approach to integrating "What-if" and LOPA ( Layer of Protection Analysis) methods. Known as a systemic safety analysis method, "What-if" can creatively identi- fy the potential hazards and likely accidents in systems or projects by means of making fault hypotheses. However, it is not likely to make quantitative hazardous potentialities and likely effective preventative measures to get rid of such dangerous events, for which reason risk assessment and risk control remains by nature unpredictable and un- surable. It is also for this reason that motivates us to introduce the LOPA (Layer of Protection Analysis) as a kind of analysis method, which helps us to effectively quantify the hazards or accidents likely to be identified by "What-if" analysis. Thus, in this paper, we would like to introduce the two methods first of all in brief, and then present an idea on how to integrate the two methods so as for them to make each up for other' s deficiencies and better fulfill the goal of risk as- sessment. In doing so, we have to put our analysis based on their re- spective functions and features and try to find the feasibility for the two methods integration through differentiating the input-output rela- tions between the two methods. And, consequently, we would like to bring about the assessment steps for the integrated method and analyze the advantages of such integration. And, last of all, demonstrate how to use the integrated method to make an authentic risk assessment on the leakage accident of toxic liquid from a container due to overpres- sure. The scenario of leakage accident will be surely identified with the effectiveness of the three Independent Protection Layers (namely, RCV, ALM + MV, SV) and the potential leakage risks enough veri- fied. The results of the example illustrate that the suggested integrated method can effectively identify the potential hazards and accidents both quantitatively and qualitatively so as to determine the size and nature