基于粒子群优化层次分析法的LNG槽车风险评价  

Risk Evaluation of LNG Tankers by Means of PSO-AHP

作　　者：李潇 张毅[1] 汪涛 李朝阳[1] LI Xiao;ZHANG Yi;WANG Tao;LI Chaoyang(School of Naval Architecture,Ocean and Civil Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学船舶海洋与建筑工程学院,上海200240

出　　处：《安全与环境工程》2022年第2期9-15,共7页Safety and Environmental Engineering

基　　金：上海市科技创新行动计划项目(20DZ1202900、21DZ1200800)。

摘　　要：液化天然气(LNG)接收站一般位于沿海主要危化品港口,港口各种生产企业密集、来往人员较多,每天往返于接收站装卸LNG的槽车,由于流动性大、安全管理细致程度高,已成为LNG接收站安全管理工作的重点。通过对槽车在LNG接收站内充装过程中全流程进行跟踪,首先选取15个影响槽车充装安全性的关键因素建立LNG接收站槽车充装风险评价指标体系;然后提出一种基于粒子群算法优化的层次分析法(即PSO-AHP法)对各风险因素的权重进行计算与重要度分析;最后针对槽车安全管理方面的薄弱点提出了相应的改进建议。结果表明:使用PSO-AHP法可以更精确地确定评价指标体系中各风险因素的权重,帮助安全管理者明确槽车充装过程中安全管理的落脚点,以增强LNG接收站内槽车运行的安全性。Liquefied natural gas(LNG)terminals are generally located in ports of hazardous chemicals,where various production enterprises cluster and people contact frequently.Due to large liquidity,safety management of high degree of detail,the safety control of LNG tankers transporting LNG every day among terminals has become the focus of safety management of LNG terminals.At present,researches on the safety of LNG tankers mainly focus on the filling technology,while macroscopic studies on safety management of the whole filling process are deficient.By tracing the whole filling process of tankers in an LNG terminal,this paper firstly selects 15 key factors affecting operation safety of LNG tankers to establish LNG terminal tanker filling risk evaluation index system,calculates the weight of each factor by the Analytic Hierarchy Process optimized by Particle Swarm Optimization(PSO-AHP),sequences these factors according to their weights and finally puts forward solutions to the problem of the system.The result shows that the weight of each factor in this evaluation system can be more accurately determined by using PSO-AHP,which can help safety managers to find foothold on safety management of the tanker filling process,and improve the safety of tankers operating in LNG terminals.

关 键 词：槽车 LNG接收站 风险评价 粒子群算法 层次分析法 

分 类 号：X913.4[环境科学与工程—安全科学]