Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
Prediction of Hourly Power Consumption for a Central Air-Conditioning System Based on Different Machine Learning Methods
Ist Teil von
Advances in Smart Vehicular Technology, Transportation, Communication and Applications, p.201-209
Ort / Verlag
Cham: Springer International Publishing
Link zum Volltext
Quelle
Alma/SFX Local Collection
Beschreibungen/Notizen
This paper uses a variety of machine learning methods to predict the hourly power consumption of a central air-conditioning system in a public building. It is found that the parameters of the central air-conditioning system are different at different times, so is the corresponding power consumption. The paper applies the time series to predict the power consumption on account of the time, which predicts the hourly power consumption based on historical time series data. Comparing the prediction accuracy of multiple machine learning methods, we find that the Gradient Boosting Regression Tree (GBRT), one of the ensemble learning methods, has the highest prediction accuracy.