An ultra-high voltage transformer health state evaluation method and system
By combining expert experience and guidelines for health status assessment, and utilizing probability density functions and health status functions, a multi-level neural network model was constructed. This solved the problems of accuracy and generalization in the health status assessment of UHV transformers, achieving high-precision assessment results and a transparent reasoning process, thereby reducing operation and maintenance costs.
CN122196864APending Publication Date: 2026-06-12CHINA ELECTRIC POWER RESEARCH INSTITUTE CO LTD +2
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA ELECTRIC POWER RESEARCH INSTITUTE CO LTD
- Filing Date
- 2026-01-23
- Publication Date
- 2026-06-12
Smart Images
Figure CN122196864A_ABST
Abstract
The application discloses a kind of extra-high voltage transformer health state evaluation method and system, the method includes: collection or with mechanism simulation method obtains fault case, constructs the first initial sample set and the second initial sample set of health state corresponding to different characteristic parameters;Establish the health degree function relationship corresponding to the different characteristic parameters, solve the health state interval corresponding to different health degree function using probability density function, construct health state sample set based on health degree function relationship;Based on the first initial sample set, second initial sample set and health state sample set, construct loss function, according to sample number division test set and verification set, train neural network model;Real-time characteristic parameter is input into trained neural network model, and the health state corresponding to each single characteristic parameter is obtained, according to the comprehensive determination rule of preestablished, the evaluation results of all characteristic parameters are fused, and the overall health state of the extra-high voltage transformer is output.
Need to check novelty before this filing date? Find Prior Art