基于IPTV的故障诊断专家系统

摘要

针对IPTV研发、测试、运行、维护人员日常的故障诊断和定位工作,开展了有关专家系统设计及其实现方法的探索性研究.研究工作主要根据IPTV系统故障诊断特点设计了基于产生式规则的知识表示方式、基于log分析描述语言对log文件进行模式匹配的算法、采用客户端/服务器的分布式结构,构建了一种基于知识的智能推理模型并进行了系统实现.实验结果表明,所建立的产生式规则模型及其软件系统,初步实现了机器自动诊断IPTV系统故障的目标,能够自动生成对人类专家有一定帮助的诊断推理报告.%Automatic diagnosis fault of IPTV system has always been considered a hard nut in expert system. This paper reports some pioneering research on a possible reasoning algorithm and its application to the fault diagnosis of IPTV system. In light of the IPTV system fault diagnosis, it designed the knowledge representation of production rule, the log analysis describe language to match the log files' pattern, the distributed structure of client/server. As shown by the tests, the system constructed on the basis of the production rule model designed in this paper is basically capable of diagnosing some simple situation of fault.