基于人工智能技术的有线传输网络故障诊断方法研究  

Research on Fault Diagnosis Method of Wired Transmission Network Based on Artificial Intelligence Technology

作　　者：黄仕雄 HUANG Shixiong(Sichuan China Mobile Communication Technology Engineering Co.,Ltd.Chengdu 611100,Sichuan)

机构地区：[1]四川中移通信技术工程有限公司,四川成都611100

出　　处：《长江信息通信》2025年第2期179-181,共3页Changjiang Information & Communications

摘　　要：为提高有线传输网络故障诊断的准确性,满足现代通信网络维护的需求。文章利用人工智能技术,提出了有线传输网络故障诊断方法研究,该方法根据数据源的类型与格式,选择相应的采集工具采集有线网络运行数据,并将其合并和集成,形成整体的数据集,利用人工智能技术提取网络故障特征,找到在故障诊断任务上表现最优的特征子集。在此基础上,构建随机森林模型,将预处理后的实时数据与特征子集输入到训练好的随机森林模型中,诊断有线传输网络故障。实验结果表明:该方法应用后,在诊断过程中未出现任何漏诊或误诊的情况,具有较高的故障诊断准确性和全面性。In order to improve the accuracy of wired transmission network fault diagnosis and meet the needs of modern communication network maintenance.This article proposes a method for diagnosing faults in wired transmission networks using artificial intelligence technology,This method selects the corresponding collection tool to collect wired network operation data based on the type and format of the data source,and merges and integrates it to form a complete dataset,Using artificial intelligence technology to extract network fault features and find the optimal subset of features for fault diagnosis tasks.On this basis,a random forest model is constructed to input preprocessed real-time data and feature subsets into the trained random forest model for diagnosing faults in wired transmission networks.The experimental results show that after the application of this method,there were no missed or misdiagnosed cases in the diagnostic process,and it has high accuracy and comprehensiveness in fault diagnosis.

关 键 词：人工智能技术 传输网络 故障诊断 随机森林模型 

分 类 号：TN929[电子电信—通信与信息系统]