摘要
针对石化设备维修费用高、可靠性低、参数难以估计的问题,提出一种基于提高因子模型的可靠性参数估计方法.应用提高因子模型描述设备的非理想维修活动,根据维修记录构造多次故障的联合分布概率密度函数,采用遗传算法搜索该函数的最大值,提出参数估计的最大似然方法,并应用于连续重整装置中.研究结果表明,装置中的氢压缩机、汽轮机和三联合加热炉是系统中的可靠性薄弱环节,而氢压缩机、换热器、水冷器、分离器、高压吸收塔以及三联合加热炉的维修效果都可以认为是理想维修.
An approach of reliability parameters estimation is proposed based on the improvement factor model in order to solve the problems of high maintenance cost, low reliability and difficult reliability parameters estimation for petrochemical equipment. The improvement factor model is applied to describe maintenance effect. The joint probability destiny function of successive failure events is proposed according to equipment maintenance records. Genetic algorithm is adopted to seek the maximum of the function to achieve the maximum likelihood estimation of the needed parameters. This approach is applied to a continuous reforming unit. The results show that the hydrogen compressor, vapor turbulence and combined furnace are the low reliability parts of this system, while the maintenance of hydrogen compressor, heat exchanger, water cooler, separator, high pressure absorption column and combined furnace can be considered as the perfect maintenances.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2008年第11期1332-1335,1371,共5页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(20676109)
关键词
参数估计
提高因子模型
非理想维修
遗传算法
石化设备
parameter estimation
improvement factor model
imperfect maintenance
genetic algorithm
petrochemical equipment
