Open AccessForecasting of daily power consumption dynamics
Author(s) -
S. Y. Petrova
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1658/1/012040
Subject(s) - box–jenkins , power consumption , power grid , series (stratigraphy) , computer science , time series , consumption (sociology) , artificial neural network , power (physics) , power network , state (computer science) , econometrics , artificial intelligence , electric power system , autoregressive integrated moving average , machine learning , mathematics , algorithm , paleontology , social science , physics , quantum mechanics , sociology , biology
The purpose of the study is to simulate the time series of power consumption in a power grid segment. It gives us be able to forecast the future state from values of the series according to current and past estimates. We want to test two different approaches: Box-Jenkins method and recurrent neural networks, for detect which of them give more accuracy and speed of prediction calculation. This article discusses the using Box-Jenkins method.