The present disclosure discloses a method for predicting spatial-temporal dynamic distribution of electric vehicle charging loads. The method and system collect load data of each charging station, denote the charging station load by a two-dimensional matrix according to a position of the charging station, and organize it into a spatial-temporal sequence; draw the charging load at each moment as a heat map, and on the basis of constructing the charging load heat map, normalizing the heat map and inputting it into the spatial-temporal dynamic prediction model for training. A void factor is added to the time dimension of the convolution kernel of the three-dimensional convolution to form a two-dimensional dilated convolution layer, such that the model can learn the information of the spatial dimension. The method can learn the spatial-temporal dynamic rules of the charging load well, so as to predict the load comprehensively in the space dimension and the time dimension. - 1/6 Loaddataofthe charging station Converting into a moment heat map Data normalization Inputting into a model for training Outputting a Adjusting prediction result hyperparameters Bad Detecting the model Good Data de-normalized utputtingafinalGresult FIGi
Method for predicting spatial-temporal dynamic distribution of electric vehicle charging loads
2021-10-28
Patent
Elektronische Ressource
Englisch
IPC: | G05B Steuer- oder Regelsysteme allgemein , CONTROL OR REGULATING SYSTEMS IN GENERAL / B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / H02J CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER , Schaltungsanordnungen oder Systeme für die Abgabe oder Verteilung elektrischer Leistung |
PREDICTING CHARGING BEHAVIOR OF ELECTRIC VEHICLE DRIVERS
Europäisches Patentamt | 2024
|Europäisches Patentamt | 2021
|