Data-driven transmission line fault location with data-efficient transfer learning

Transfer of learning Line (geometry)
DOI: 10.1049/icp.2023.2408 Publication Date: 2023-11-13T20:09:02Z
ABSTRACT
Transmission line fault location is one of the essential steps to ensure power supply reliability. Traditional model based methods and traveling wave have limitations such as requirements accurate parameters or high sampling rates. Existing data-driven usually require large number training data that are exactly consistent with practical system. However, in systems quite limited, there could be mismatch between system simulation system, limiting accuracy. To this end, paper proposes a transfer learning method for transmission lines. The can efficiently utilize small dataset systems. First, neural network constructed pre-trained extensive generated by A. Next, another very B mimic scenario, where different from utilizes update network, freeze-training fine-tuning. Finally, performances without compared. results clearly indicate effectiveness necessity proposed method.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (0)
CITATIONS (0)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....