Real-time simulation system of low-voltage meter collection network full-link communication state based on digital twin NS3–MATLAB co-simulation

With the deepening of global energy transformation, the scale and complexity of smart grids are rapidly increasing. Low-voltage power line communication networks, as key infrastructure for connecting massive end-users, face significant challenges in communication reliability. To address these systematic issues, this study proposes a co-simulation architecture based on the deep integration of digital twin and link quality awareness. The framework enables seamless cross-platform collaboration between NS-3 and MATLAB. By implementing a dynamic route convergence mechanism, the average convergence time is optimized from 382.5 to 145.8 ms, achieving a 62% performance improvement. The system also attains a high self-healing success rate of 99.85% and enhances throughput to 1.52 Gbps, which is 45% higher than traditional approaches. Experimental results confirm the robustness of the proposed scheme in highly dynamic scenarios while supporting simulation and evaluation for new service deployment, network expansion, and extreme failure conditions. This study provides a reliable quantitative analysis framework for precise planning and proactive operation and maintenance of communication networks.