Akademska digitalna zbirka SLovenije - logo
(UM)
  • Support of machine learning in assessing distribution network consumer flexibility potential [Elektronski vir]
    Rober, Jaka ; Beković, Miloš ; Maruša, Leon
    This paper presents a step-by-step approach to assessing the flexibility potential of residential consumers to manage congestions. A case study is presented where a selected transformer station ... exhibits signs of overloading. Based on historical load data, analysis has been made to evaluate the magnitude of overloading and timing of overload occurrence. The four most prominent consumers have been chosen for flexibility assessment based on historical load data. Machine learning algorithms, specifically multiple linear regression and support vector machines were employed for load profile forecasting during overload occurrences. The generated models were evaluated and compared with forecasting based on the average load of the past days. Based on the evaluated profiles, a scenario of flexibility has been made for each consumer that has been selected as having highest potential for flexibility services. The results demonstrate the effectiveness of the machine learning models, which outperform the averagebased forecasting method and provide more realistic estimates of flexibility potential. The proposed approach can be applied to other overloaded transformer stations but with a limited number of consumers.
    Source: Zbornik dvaintridesete mednarodne Elektrotehniške in računalniške konference ERK 2023 [Elektronski vir] = Proceedings of the 32nd International Electrotechnical and Computer Science Conference ERK 2023 : Portorož, Slovenija, 28. - 29. september 2023 (Str. 469-472)
    Type of material - conference contribution ; adult, serious
    Publish date - 2023
    Language - english
    COBISS.SI-ID - 167950595

    Link(s):

    https://erk.fe.uni-lj.si/2023/papers/rober(support_of).pdf

Shelf entry