中文摘要：

针对模糊综合评价法在铁路时间同步网可靠性评价中的不足,提出了基于组合赋权的铁路时间同步网云评估模型。首先,根据铁路时间同步网的功能结构,构建系统综合能力模型,建立其对应的功能硬件层次结构,从而选取影响铁路时间同步网可靠性的主要评价指标。然后,利用标度扩展AHP法和熵权法分别计算铁路时间同步网各评价指标的主客观权重,并且通过引入欧几里德距离函数计算差异系数,使主、客观权重之间的偏离程度与权重系数一致,从而获得各指标合理的综合权重。最后,计算各指标相对值,并且利用云模型表示,进行云计算得到综合指标云,将该综合指标云与可靠性评价集的云模型进行相似度度量,得到铁路时间同步网综合可靠性评价结果。结果表明,基于组合赋权的铁路时间同步网云评估模型能够实现对铁路时间同步网综合可靠性的评价。

英文摘要：

This paper is intended to introduce a reliability assessment method based on the cloud model and the integrated weight to evaluate the railway time control synchronization network. The reason and the initiative for our research come from the starting point that most of the current methods for assessing the system reliability are to focus largely on the scoring the policy assessed by the experts in the corresponding areas. In so doing,the evaluation results fail to come well in accord with the objectively existing situation or situations. For example,the railway time synchronization network evaluation can serve as a typical complicated hierarchical functional system, the traditional two-state system of which would fail to come to the need of such complicated reliability evaluation objectives. Moreover,the said method can neither reflect the actual situation of the railway time synchronization network accurately. It is just coming from the inadequacy of the above mentioned methods that this paper would like to propose a more comprehensive and reliable method for the reliability evaluation here based on the combination of weighting with the cloud model. In doing so,first of all,it is necessary to build up a comprehensive model of the railway time synchronization network on the basis of its structure and function through the analytic hierarchy process with the evaluation indexes being chosen for the necessary analysis. And, secondly, the subjective and objective weights of the related parameters are to be worked out by the integrated use of the scale-extending AHP and the entropy weight method. That is to say,by introducing the Euclidean distance function,it would be possible to work out the differences between the subjective and objective weights,for they would be the same as the deviation between the different coefficients,as well as the reasonable combination weights. Thus,eventually,it would be possible to work out the relative values to be expressed by the cloud model and by the cloud operating rules. The cloud