基于检验测试的安全仪表系统安全评价Markov简化模型

Simplified Markov Model for Safety Evaluation of Safety Instrumented System Based on Proof Tests

Markov模型是研究安全仪表系统(Safety instrumented system,SIS)安全性的重要模型之一,但随着SIS运行特性日益复杂,其状态数量呈指数性增长,导致计算量剧增。考虑SIS运行阶段检验测试的特性,通过计算状态间转换率、确定主导状态、合并相似状态、引入修正因子补偿计算偏差等方法对Markov模型进行了简化,提出了一种系统的基于检验测试的要求时危险失效平均概率(PFDavg)的Markov简化计算模型,适用于低要求运行模式下同构冗余架构系统的SIS,该模型保留了Markov原始模型的主要特性,验证结果表明:在安全参数置信区间内,该模型的计算复杂度显著降低,结果误差在5%内,不会影响决策。

Markov model is one of the most important models for the study of the Safety Instrumented System(SIS).However,with increasingly complex operation characteristics of the SIS,the number of Markov states has experienced exponential growth.Thus,the Markov model faces the challenge of enormous computation amount while analysing SIS safety.This work considers test characteristics in the SIS operation stage and simplifies the Markov model by calculating state transition rate,determining the dominant state,merging similar states and introducing correction factors to compensate calculation deviation.With these ways of simplication,this paper proposes a test-based simplified Markov calculation model of the average Probability of danagerous Failure on Demand(PFDavg)for the homogeneous reduntant SIS system.The proposed model maintains major characteristics of a Markov model.The verification results show that the calculation amount of the presented model is obviously smaller than that of the original Markov model,and the calculation error is smaller than 5%within the confidence interval of the safety parameters.Thus,the proposed model is reliable enough to assist decision-making.