结合模糊网络环境特征，提出了在全局网络中遍历服务异常（QoS）的目标的评估函数。该函数的实施，首先 通过建立节点的数学模型，在此基础上分析传统研究方法开展评估过程中表现出的局限性和适应性。其次，结合网络节点多 跳特征给出相应机制筛选信息域。同时引入指数级评估函数用于排除累积性差错引发的跳距误差。然后形成一套评估函数 的实施策略。最后，搭建实验模型部署实验方案，将评估函数同传统研究方法开展对比。经过三项实验的多次考察表明，本文 研究的指数级评估函数部署在模糊网络中，各项指标均占据优势。

Based on the characteristics of fuzzy network environment, this paper presents an evaluation function for traversing the target of Question of Service (QoS) in global network. In the implementation of the function, firstly, the mathematical model of the node is established, and then the limitations and adaptability of the traditional research methods in the evaluation process are analyzed. Secondly, combined with the multi-hop features of network nodes, the corresponding mechanism is given to filter the information domain. At the same time, an exponential evaluation function is introduced to eliminate the jump error caused by cumulative errors. Then it forms a set of evaluation function implementation strategy. Finally, the experimental model is constructed and the evaluation function is compared with the traditional research method. The results of three experiments show that the exponential evaluation function in this paper is deployed in the fuzzy network and each index occupies the advantage.