MODEL TRAINING METHOD, POWER PREDICTION METHOD, AND DEVICE

The present application discloses a model training method, a power prediction method, and a device, which can be applied to the field of artificial intelligence. A model at least comprises a first model and a second model. The method comprises: acquiring historical power data and historical meteorol...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: RAO, Zhongwen, PAN, Lujia
Format: Patent
Sprache:chi ; eng ; fre
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The present application discloses a model training method, a power prediction method, and a device, which can be applied to the field of artificial intelligence. A model at least comprises a first model and a second model. The method comprises: acquiring historical power data and historical meteorological data in preset durations; determining weights of loss functions of the models on the basis of a data set and constructing a target loss function, wherein the first model and the second model can be used for representing input data, and the data represented by the first model and the data represented by the second model are not completely the same; and finally, training the models by using the target loss function. By using a currently available data source, more complete information is captured from different modal data (i.e., historical power data and historical meteorological data) and different time scales (i.e., preset durations), so that a learning model is optimized, and higher prediction precision can