水电机组保护装置故障样本的生成方法及相关设备
By preprocessing historical operating data and constructing an improved generative adversarial network model framework, high-quality and diverse fault sample data are generated, solving the problem of scarce and unbalanced fault samples in hydropower unit protection devices and improving the accuracy and reliability of fault diagnosis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUBEI ENERGY GRP CO LTD
- Filing Date
- 2026-06-15
- Publication Date
- 2026-07-17
AI Technical Summary
Fault sample data for hydropower unit protection devices are scarce and lack diversity. Existing technologies generate samples with low accuracy to real fault characteristics, and the training process is unstable, making it difficult to meet the needs of fault diagnosis.
By preprocessing historical operational data, a generative adversarial network model framework is constructed that integrates a dynamic adaptive gradient penalty mechanism, historical fault distribution constraints, and an adaptive early stopping strategy based on the F1 stability index, thereby generating high-quality and diverse fault sample data.
The generated fault sample data closely matches the real fault characteristics, solving the problems of sample scarcity and uneven distribution, improving the accuracy and reliability of the fault diagnosis model, and ensuring the safe and stable operation of the hydropower unit protection device.
Smart Images
Figure CN122412964A_ABST