Intelligent error-proof logic multi-factor cooperative active optimization based on knowledge base and generative adversarial network
In order to solve the reduced prediction accuracy caused by insufficient training samples, data imbalance and improper feature selection of the intelligent error prevention system in the power grid, an intelligent errorproof logic multi-factor cooperative active optimization based on a knowledge bas...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Systems Science & Control Engineering |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/21642583.2025.2486126 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In order to solve the reduced prediction accuracy caused by insufficient training samples, data imbalance and improper feature selection of the intelligent error prevention system in the power grid, an intelligent errorproof logic multi-factor cooperative active optimization based on a knowledge base and generative adversarial network was designed. Combining the logic rules given by the power system and the real-time operation parameters of the substation, the intelligent terminal is used to monitor the remote control information of the anti-misoperation latching device in real-time and the power equipment is controlled by remote control of the anti-misoperation latching device. Attribute reduction of the decision table uses a niche genetic algorithm to extract the intelligent anti-failure logic rule set. The generative adversarial network is applied to the distribution network error-proof logic, the discriminator and generator are trained alternately, the data generated by the adversarial network are used to enrich the knowledge base, and then the rules in the knowledge base are used to guide the fault diagnosis process. Experiments show that the design method has a real-time data refresh response time of 80 to 85 ms, a peak accuracy of 98.66% and a minimum electrical loss rate of 0.42%. |
|---|---|
| ISSN: | 2164-2583 |