基于模糊推理的FRAM事故定量分析方法研究

为有效克服FRAM事故分析中无法进行定量分析的缺陷，提出结合模糊推理技术的Fuzzy FRAM模型。此改进模型基于FRAM识别系统运行状态；依据功能输出要素的时间/精度属性利用Matlab构建2阶模糊推理系统量化功能输出质量；根据通用性能条件(CPC)及功能输入耦合端口构建功能评价体系，针对评价体系中存在的不确定性信息融合及建模问题，采用模糊证据推理技术，通过模糊信度结构建立、数据处理、信息融合测度后获得功能的风险指数;以既有铁路危险品运输事故为例，验证方法的可行性。结果表明:Fuzzy FRAM模型的评估结果较为精确，是FRAM分析方法的有效补充。

Research on FRAM accident quantitative analysis method based on fuzzy reasoning

In order to effectively overcome the defect that the quantitative analysis can’t be conducted in the FRAM (functional resonance analysis method) accident analysis,a Fuzzy FRAM model combined with fuzzy reasoning technology was proposed.The improved model accomplished the recognition on the operation state of system based on FRAM,and according to the time/precision properties of output elements,a second order fuzzy inference system was built in MATLAB to obtain the quantitative evaluation of functional output.Then,the function evaluation system was constructed according to the common performance condition (CPC) and function input coupling port.In view of the uncertain information fusion and modeling problems in the evaluation system,the fuzzy evidence reasoning technology was used to obtain the risk index of functions after the establishment of fuzzy reliability structure,data processing and information fusion measurement.Finally,the feasibility of the method was verified by taking a railway dangerous goods transportation accident as an example.The results showed that the evaluation results of Fuzzy FRAM model was more accurate,which is the effective supplement to FRAM analysis method.