abstract |
The invention relates to the field of hydropower station data management, and aims to provide a data prediction model subdivision method based on the state of hydropower station equipment, applying machine learning to construct a model conforming to the operation law of hydropower generator set equipment, including known faults, normal operation, For regular events and occasional events, according to the requirements of continuous data characteristics of products, the data is enriched in time series, which provides effective support for enriching modeling factors. |