通信学报2025，Vol.46Issue(4)：255-271,17.DOI:10.11959/j.issn.1000-436x.2025064
面向自智算力网络的数字孪生:架构与关键挑战
Digital twin for self-intelligent computing power networks:architecture and key challenges
摘要
Abstract
Current self-intelligent computing power networks were challenged by delayed responses to complex failure scenarios,inefficient deployment of optimization strategies,and massive data operation dilemmas.To address these is-sues,virtual replicas were constructed through digital twins,which achieved real-time monitoring and failure deduction,enabling intelligent operation and maintenance to facilitate breakthroughs in development bottlenecks.Based on the defi-nitions of self-intelligent computing power networks and digital twins,a digital twin architecture targeting such networks was proposed.Design principles,deployment methodologies,and typical application scenarios were systematically elaborated,while key challenges in applying digital twin technology were analyzed through case studies.It demonstrates that digital twins effectively enhance network autonomy and operational efficiency.Finally,several open issues regard-ing model fidelity and interoperability are identified,and future research directions for digital twin applications in self-intelligent networks are outlined.关键词
自智算力网络/数字孪生/资源管理/智能决策Key words
autonomous computing power network/digital twin/resource management/intelligent decision-making分类
信息技术与安全科学引用本文复制引用
黄韬,周子翔,唐琴琴,谢人超..面向自智算力网络的数字孪生:架构与关键挑战[J].通信学报,2025,46(4):255-271,17.基金项目
国家重点研发计划基金资助项目(No.2024YFB2906701) The National Key Research and Development Program of China(No.2024YFB2906701) （No.2024YFB2906701）
