Substation ground oil pollution identification method, device, electronic equipment and storage medium
Through the semantic recognition model of the full convolution network and residual network combined with the random weight averaging method, the problem of slow recognition speed and low accuracy of ground oil pollution in the substation is solved, and automatic identification and timely feedback of ground oil pollution in the substation is realized, which improves the recognition accuracy and equipment safety.
Patent Information
- Application Number
- CN202210557795.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-19
- Publication Date
- 2025-08-26
- Estimated Expiration
- 2042-05-19
AI Technical Summary
In the prior art, the identification of ground oil pollution in substations relies on manual detection, which is slow, poor real-time and low accuracy, making it difficult to detect oil leakage problems in a timely manner, affecting the life of the equipment and the safety of the power system.
The ground oil-film semantic recognition model based on full convolutional network and residual network is adopted, and the model parameters are trained in combination with the random weight averaging method, and the oil-film area and oil leakage degree are automatically identified and output. The semantic segmentation network combines shallow position information and deep semantic information to improve the recognition accuracy and speed.
It realizes efficient automatic identification of oil pollution on the substation, improves identification accuracy and speed, promptly feedback on oil leakage, supports rapid maintenance, and improves the safety and equipment life of the substation.