Abstract: This study proposes a data-driven resilience planning method, establishes a full-cycle management system for intelligent equipment, and develops elastic reconfiguration technology based on multi-source integration, thereby forming a closed-loop approach from problem diagnosis to strategy implementation. The results show that the dynamic planning model can reduce load forecasting errors by 24%, while the network reconfiguration technology can enhance renewable energy consumption capacity by 18%. These findings provide both theoretical support and practical pathways for the construction of new-type power systems.