基于模糊理论的PMT水下防爆试验故障树分析

为解决光电倍增管（photomultiplier tubes，简称PMT）水下防爆试验系统的故障诊断问题以及故障发生概率存在模糊不确定性的特点，应用模糊集数学方法和可能性理论，把模糊数学和故障树模型相结合，提出一种基于三角形模糊数算法的PMT试验系统失效模式分析方法。根据试验系统工作原理，采用下行法建立故障树并对其进行定量分析，获得了顶事件的结构函数及多种失效模式，进而利用专家语言评价法，计算顶事件发生的模糊概率和各底事件的模糊重要度，得到了PMT试验中故障发生率较高的环节，最后对故障分析结果进行改进测试。结果表明，该方法能够准确地诊断出存在的故障，为PMT防爆试验系统进行故障诊断以及提高PMT保护罩的水下防爆性能提供参考依据。

Fuzzy Fault Tree Analysis of Underwater PMT Explosion Protection Experiment System

To estimate the failure probability of the photomultiplier tubes (PMT) underwater explosion test system, a method using information from fault tree analysis and uncertainty of data is presented. Based on the fuzzy set method and possibility theory, fuzzy fault tree method is a mathematical approach to model inaccuracy and uncertainty of the failure. The proposed model is able to quantify the fault tree of PMT explosion protection experiment system. In order to effectively investigate basic event failure probabilities without crisp quantitative historical failure data, fuzzy logic theory combined with expert judgement method is employed. The failure probability of the top event and the importance measures of basic events is also illustrated. The result demonstrates that the approach is apposite for the assessment of system failure. The research results can help professionals to decide whether and where to take preventive or corrective actions and help improve the underwater explosion-proof performance of PMT protective cover and fault diagnosis of test system.