A tie line planning method and system based on deep learning
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA ELECTRIC POWER RESEARCH INSTITUTE CO LTD
- Filing Date
- 2020-05-07
- Publication Date
- 2026-07-24
AI Technical Summary
The different planning methods used by dispatch centers at various levels make it impossible to effectively grasp the operation mode and planning calculation methods of power grids at all levels, resulting in repeated iterations and modifications to tie line plans.
A deep learning-based long short-term memory network model is adopted. By training historical load data, clean energy power data and tie-line planned power data, a mapping relationship between system load, clean energy power and tie-line planned power is established, and tie-line plans are directly generated using prediction data.
It improves the decision-making accuracy and adaptability of tie-line planning, reduces iterative modifications, and has the characteristics of self-evolution and self-learning, making it suitable for computational problems in complex interconnected large power grids.
Smart Images

Figure CN111709550B_ABST