基于改进DSmT的公路隧道行车安全综合评价方法研究

Comprehensive evaluation of highway tunnel traffic safety based on improved DSmT

为解决公路隧道行车安全综合评价中存在的模糊性、不准确性与冲突性问题,将综合评价视作一个以评价指标值为信息源的融合过程,提出基于改进DSm T的综合评价方法,兼顾评价指标的模糊性和约束性,优化评价结果。首先,利用云模型云化安全等级,使其尽可能连续,降低以安全等级为鉴别框架的各命题的约束能力,提高信息的准确性;其次,用云模型处理定性评价指标自然语言值与量化数值间的转换问题,进行基于隶属云的信息源广义基本信度赋值(GBBA),实现多级证据体的信息融合;最后,以京台高速福建段牛岩山隧道为例,进行工程实例分析。结果表明,用该评价方法能够处理定性指标量化问题,满足指标体系的动态要求,评价结果与实际相符。

The paper was aimed at solving the problem of fuzziness, uncertainty and conflict in comprehensive evaluation of expressway tunnel traffic safety. Comprehensive evaluation was regarded as a fusion process taking the evaluation index value as information source. Then a new method based on improved DSm T theory was worked out for optimizing the evaluation results, considering both the fuzziness and the restrainability of evaluation indexes. Firstly, cloud model was used to make safety grade cloud and make it as continuous as possible to reduce the constraints of proposition of framework. Secondly, cloud model was used to convert natural language value to quantitative value, and realize GBBA of information sources based on membership cloud. Finally, the evaluation method was verified by a project example of Niuyanshan tunnel in Fujian section of Beijing Taiwan expressway. The results show that the method can be used to deal with quantification of qualitative index, meet the dynamic requirements of index system, and that the evaluation conforms with the actual situation.