传统的故障树多用于定性分析,在定量分析时主观随意性较大。根据专家经验确定底事件的发生概率,通过三角模糊数和算子公式得到顶事件的发生概率,对故障树各层故障指标进行专家层次分析和评估,基于聚类法剔除偏离度严重的数据,进行综合评价。最后开发了聚类模糊综合评价的原型系统,并以船舶柴油机排放烟色异常为实例,分析并建立故障树,实现了船舶柴油机排放烟色的聚类模糊综合评价。
The traditional fault tree are usually used for some qualitative analysis,the subjective and arbitrary is large in quantitative analysis. This paper determined the probability of occurrence of events based on expert experience,and got the probability of occurrence of the top event of the fault through the triangular fuzzy number and operator formula. This paper used expert analysis and assessment of the degree of deviation for fault indicators of every hierarchical layers of the fault tree, excluding serious data based on clustering method and using fuzzy comprehensive evaluation. Finally,this paper developed a clustering fuzzy comprehensive evaluation of the prototype system,and used the failure of marine diesel engine that emits abnormal smoke as an example, it also established the fault tree of marine diesel engine that emits abnormal smoke based on clustering fuzzy comprehensive evaluation.