Small sample user non-intrusive load monitoring method based on transfer learning
The invention relates to a small sample user non-intrusive load monitoring method based on transfer learning, and the method comprises the steps: obtaining reference load data of a plurality of electric devices, the reference load data comprising reference total energy consumption data of all electr...
Gespeichert in:
Hauptverfasser: | , , , , , , , , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention relates to a small sample user non-intrusive load monitoring method based on transfer learning, and the method comprises the steps: obtaining reference load data of a plurality of electric devices, the reference load data comprising reference total energy consumption data of all electric devices; dividing the reference load data into public data set-based source domain data and actually measured residential user data-based target domain data; inputting the source domain data into a deep neural network for pre-training, and performing feature extraction on the source domain data to obtain a complex spatial-temporal feature relationship between loads; and migrating the pre-trained model parameters to the target domain to realize load monitoring of the target domain. The load monitoring result is the on-off state of each electric device, and the power utilization rule of all electric devices is analyzed according to the load monitoring result. Compared with the prior art, the method has the advanta |
---|