24-Hours Demand Forecasting Based on SARIMA and Support Vector Machines

Prévision de la demande sur 24 heures fondée sur des modèles SARIMA et SVM

Mathias Braun (1) , Thomas Bernard (1) , Olivier Piller (2) , Fereshte Sedehizade (3)

1 Fraunhofer IOSB - Fraunhofer Institute of Optronics, System Technologies and Image Exploitation
2 UR ETBX - Environnement, territoires et infrastructures
3 BWB - Beerliner Wasserbetriebe

In time series analysis the autoregressive integrate moving average (ARIMA) models have been used for decades and in a wide variety of scientific applications. In recent years a growing popularity of machine learning algorithms like the artificial neural network (ANN) and support vector machine (SVM) have led to new approaches in time series analysis. The forecasting model presented in this paper combines an autoregressive approach with a regression model respecting additional parameters. Two modelling approaches are presented which are based on seasonal autoregressive integrated moving average (SARIMA) models and support vector regression (SVR). These models are evaluated on data from a residential district in Berlin.

Mots clés: MODEL CALIBRATION; WATER DISTRIBUTION SYSTEM

Mots clés: DEMANDE EN EAU; CALAGE DE MODELE; PREDICTION; RESEAU DE DISTRIBUTION D'EAU

Domaines: Machine Learning [stat.ML]; Mathématiques [math]; Sciences de l'ingénieur [physics]

Mathias Braun, Thomas Bernard, Olivier Piller, Fereshte Sedehizade. 24-Hours Demand Forecasting Based on SARIMA and Support Vector Machines. Procedia Engineering, 2014, 89, pp.926-933. ⟨10.1016/j.proeng.2014.11.526⟩. ⟨hal-01132007⟩