Akademska digitalna zbirka SLovenije - logo
VSE knjižnice (vzajemna bibliografsko-kataložna baza podatkov COBIB.SI)
  • Data driven classification and forecasting of residential load profiles [Elektronski vir]
    Markovič, Rene ; Marhl, Marko ; Virtič, Peter
    Smart meter systems allow us to follow and record the energy consumption behavior of individual households. The gathered data can then be used to study the consumption habits of individual user, ... clusters of users or a global bulk of user. In addition to gaining knowledge about the consumption habits we can develop mathematical models, which behave in a similar way the user does. Such models are used to make predictions, which are needed for the proper long-term planning of transmission and distribution networks as well as short-term scheduling and security functions of an energy management system. Herein we study how different data aggregation techniques influence forecasting accuracy of three data driven forecasting models. Our results give new insights about the factors that determine the accuracy of forecasting models and their forecasting potential.
    Vir: Digital proceedings [Elektronski vir] (12 str.)
    Vrsta gradiva - prispevek na konferenci ; neleposlovje za odrasle
    Leto - 2018
    Jezik - angleški
    COBISS.SI-ID - 1024329564

Vnos na polico