计算单点滑环系统失效概率的蒙特卡洛方法  

A Monte-Carlo Method to Calculate the Failure Probability of a Single Point Swivel System

作　　者：何鑫[1] 唐友刚[1] 付礼鹏 孙丽萍[2] 刘骄 HE Xin;TANG Yougang;FU Lipeng;SUN Liping;LIU Jiao(School of Civil Engineering, Tianjin University, Tianjin 300354, China;School of Shipbuilding Engineering, Harbin Engineering University, Harbin 150001, China)

机构地区：[1]天津大学建筑工程学院,天津300354 [2]哈尔滨工程大学船舶工程学院,哈尔滨150001

出　　处：《船舶工程》2018年第4期82-88,共7页Ship Engineering

基　　金：工信部"FPSO失效数据库及风险评估系统研究"(G014614002)

摘　　要：建立单点滑环系统失效的动态故障树模型,利用齐次马尔科夫链表示状态空间的转移过程,并求出滑环系统失效概率精确解。提出计算单点滑环系统失效概率的蒙特卡洛方法,编制蒙特卡洛仿真程序,求解齐次马尔科夫链的数值解,结果显示,当抽样数等于5 000时,结果趋近收敛,数值解与精确解吻合良好。对于非齐次马尔科夫链,由于无法求出精确解,因此利用蒙特卡洛法求解滑环系统失效概率的数值解,数值解失效概率的变化规律与前述结果一致,验证了蒙特卡洛法的适用性。利用蒙特卡洛法求解系统停产的概率,与滑环系统失效概率进行对比发现,在相同工作时间下,系统停产概率明显小于系统失效概率。结果显示,蒙特卡洛法的收敛性好,结果精度高,且能解决马尔科夫法不能解决的问题,可以很好地应用在单点滑环系统的风险评估中。A dynamic fault tree is formed for the single point swivel system, a homogeneous Markov chain is adopted to express the transferring process between each of the states, and an exact solution of the swivel system’s failure probability is obtained. A Monte-Carlo method is brought forward to calculate the failure probability of the single point swivel. A Monte Carlo simulation program is developed to numerically solve the homogeneous Markov chain. The results show that the numerical solution tends to converge and the numerical solution fits well with the exact solution with a sample number of 5000. For a non-homogeneous Markov chain, the exact solution cannot be obtained, so that the Monte-Carlo method is used to solve the numerical solution of the failure probability of the swivel system. The variation law of the failure probability showed in the numerical solution is in accordance with the former one, which verifies the applicability of the Monte-Carlo method. The Monte-Carlo method is used to calculate the probability of system shutdown and the result is compared with the failure probability of swivel system. It shows that in the same service time, the probability of the system shutdown is obviously lower than the system failure probability. The results show that the Monte-Carlo method converges well, is of high precision, and can solve problems that the Markov method cannot. The Monte-Carlo method is well applicable in the risk assessment of the single point swivel system.

关 键 词：单点系泊 滑环 动态故障树 马尔科夫链 蒙特卡洛仿真 风险评估 

分 类 号：U661.43[交通运输工程—船舶及航道工程]