Prophet-Gru model-based large hydraulic turbine set bearing bush temperature prediction method
By using the Prophet-Gru model in the hydraulic turbine unit to predict the bearing tile temperature in combination with historical data and various influencing factors, the problem of single, cumbersome and lagging prediction methods in the existing technology is solved, and higher prediction accuracy and more accurate predictions are achieved. Early fault detection.
CN119358722AInactive Publication Date: 2025-01-24CHINA YANGTZE POWER
Patent Information
- Application Number
- CN202411256869.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2025-01-24
- Estimated Expiration
- Not applicable · inactive patent
Smart Images
Figure CN119358722A_ABST
Abstract
The invention relates to a Prophet-Gru model-based bearing bush temperature prediction method for a large hydraulic turbine set, which comprises the following steps of: preprocessing historical data of four parts, namely an upper guide bearing bush temperature, a lower guide bearing bush temperature, a thrust bearing bush temperature and a water guide bearing bush temperature, and corresponding set active power, cooler group number, oil level of an oil tank, air temperature and oil temperature data, and respectively establishing a Prophet model and a Gru model; wherein the Prophet model can better reflect the time sequence characteristics of the bearing bush temperature, the Gru model fully reflects the influence of each influence factor on the bearing bush temperature, then the Prophet model and the Gru model are summed to reconstruct a new model Prophet-Gru model, analysis is carried out in combination with model evaluation indexes, and therefore a predicted value is obtained, and compared with a traditional prediction method, the method has better accuracy.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Method for predicting temperature of guide bearing bush of water-turbine generator set through machine learning
CN112560339A
User electricity consumption prediction method based on Prophet-LSTM model
CN112990556A
Prophet-BIGRU-based gas daily load prediction method
CN115689036A
Water turbine bearing bush temperature prediction method based on machine learning and neural network
CN116542121A
Water demand prediction method and system
CN117610826A