一种多电站协同缺失数据插补方法及系统
By using a multi-power station collaborative missing data imputation method, dynamically selecting the optimal neighbor, combining adaptive graph learning and dynamic graph attention mechanisms, and utilizing TimesNet periodic folding technology, the problems of low computational efficiency and insufficient information utilization in multi-power station missing data repair are solved, achieving high-precision imputation results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV OF SCI & TECH
- Filing Date
- 2026-05-19
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies suffer from low computational efficiency, weak periodic modeling capabilities, and insufficient utilization of information from multiple power plants when processing missing data from multiple power plants, thus failing to achieve accurate missing data repair.
A multi-power station collaborative missing data imputation method is adopted. The optimal neighbor set is selected through a dynamic multi-site selection mechanism. Combined with adaptive graph learning and dynamic graph attention mechanism, spatiotemporal features are explicitly extracted using TimesNet periodic folding technology for imputation.
It significantly improves interpolation accuracy and computational efficiency, effectively utilizes information from multiple power stations, dynamically adjusts spatiotemporal weights, and ensures the physical authenticity of periodic features and the accuracy of interpolation results.
Smart Images
Figure CN122220711B_ABST