Digital twin for self-intelligent computing power networks: architecture and key challenges

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 issues, 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 definitions 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 regarding model fidelity and interoperability are identified, and future research directions for digital twin applications in self-intelligent networks are outlined.