Fully self-developed microgrid digital twin engine, with embedded AI-driven causal reasoning engine, improves microgrid self-healing response speed by 4.8 times on average, reduces manual intervention frequency by 91.3%
PDT's independently developed microgrid digital twin engine has been deployed in 37 actual project projects nationwide. The engine supports millisecond-level topology dynamic reconstruction and multi-time-scale simulation deduction, with an embedded AI-driven causal reasoning engine that can autonomously identify root causes of source-load-storage coupling imbalance and generate executable closed-loop control strategies.
Its decision-making process is fully traceable, explainable, and auditable, ensuring each instruction originates from real physical constraints and economic trade-offs. To date, the engine has achieved:
- Fault prediction accuracy ≥92%
- Strategy iteration cycle compressed to within 24 hours
- Average improvement of microgrid self-healing response speed by 4.8 times
- Reduction of manual intervention frequency by 91.3%