Probabilistic Forecasting of Electricity Demand Incorporating Mobility Data

Due to extreme weather conditions and anomalous events such as the COVID-19 pandemic, utilities grid operators worldwide face unprecedented challenges. These unanticipated changes in trends introduce new uncertainties conventional short-term electricity demand forecasting (EDF) since its result depends on recent usage an input variable. In order quantify uncertainty of EDF effectively, this paper proposes a comprehensive probabilistic EFD method based Gaussian process regression (GPR) kernel density estimation (KDE). GPR is non-parametric Bayesian theory, which can handle using limited data. Mobility data incorporated manage pattern increase model scalability. This study first performs correlation for feature selection that comprises weather, renewable non-renewable energy, mobility Then, different functions are compared, optimal function recommended real applications. Finally, used validate effectiveness proposed elaborated with three scenarios. Comparison results other adopted methods show achieve high accuracy minimum quantity while addressing uncertainty, thus improving decision-making.