贝叶斯方法在安全仪表系统设备失效率确定中的应用

Application of Bayesian method in determining safety instrumented system equipment failure rate

为获得更符合企业实际的安全仪表系统(SIS)设备失效率,应用贝叶斯方法,以海上平台可靠性数据手册(OREDA.)数据库信息作为设备失效率先验信息,综合先验信息、企业收集的相关设备样本信息和蒙特卡洛方法仿真的企业样本信息,得到设备失效率贝叶斯估计。并以某石化装置为例,分别将设备失效率的贝叶斯估计和数据库信息代入不同模型进行安全完整性等级(SIL)验证并对比。结果表明,通过贝叶斯方法获得的设备失效率更为合理,使系统由SIL3降为SIL2。

To obtain more practical reliability data handbook （OREDA） equipment failure rate of SIS, Bayesian method was used. Offshore information was used as the prior information of equipment failure rate, then Bayesian estimation of equipment failure rate is obtained after integrating the prior information, enterprise related equipment sample information and Monte Carlo simulation of enterprise sample information. Taking a certain petrochemical plant as an example, the Bayesian estimation of equipment failure and database information were respectively substituted into different models to verify SIL and compare the results. The comparison shows that equipment failure rate obtained by Bayesian method is more reasonable and the SIS can be degraded from SIL3 to SIL2.