Method and System for Dynamic Load Balancing Optimization of Ring Main Unit
By combining the construction of admittance matrix within the ring network box with an improved graph neural network, the problem of asynchronous timing error in measurement data in medium-voltage distribution networks was solved, achieving efficient dynamic load balance optimization and improving calculation accuracy and response speed.
Patent Information
- Application Number
- CN202611141265.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-30
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2046-07-30
AI Technical Summary
Existing technologies in medium-voltage distribution networks suffer from asynchronous timing errors in measurement data due to differences in wireless communication environments. This causes the master station's calculation results to deviate from the actual power grid state, making it difficult to meet the high timeliness requirements of dynamic balance control under scenarios of severe load fluctuations.
By constructing an admittance matrix within the ring network box, performing singular value decomposition, extracting left singular vectors to form a topological matrix, and using an improved graph neural network to estimate the time delay factor, a confidence matrix is generated to filter and reconstruct the load current vector. The data is then sent to the cloud optimizer to calculate the interconnection switch state change strategy only when the confidence level meets the standard.
It effectively suppresses time misalignment measurement distortion caused by communication latency differences, improves the accuracy and response speed of dynamic load balancing, and avoids erroneous control based on distorted data.
Smart Images

Figure CN122678005B_ABST
Abstract
Citation Information
Patent Citations
Intelligent power grid management method based on electric energy metering box
CN121961114A
Fault root cause analysis method fusing graph neural network and equipment topological relation
CN122387729A
