Open AccessNeural network in a “sliding window” for power grids signals structural analysis
Author(s) -
Oleg N. Andreev,
Alexandr L. Slavutskiy,
Leonid A. Slavutskii
Publication year - 2022
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/990/1/012054
Subject(s) - artificial neural network , transformer , voltage , current transformer , computer science , signal (programming language) , electrical engineering , electronic engineering , electric power system , current sensor , engineering , power (physics) , artificial intelligence , physics , quantum mechanics , programming language
In electric power systems, intelligent electronic devices monitor the state of an energy facility in real time. The data for monitoring the quality of electricity at the intelligent electronic device is transmitted through current and voltage measuring transformers. The operating modes, when the power transformer is turned on under voltage, as well as in emergency modes, signal distortions occur in the transformer. This paper is concerned with simplest feed forward artificial neural network to estimate the parameters of a distorted signal. The neural network is used to estimate the parameters of the current signal in the secondary winding of the current measurement signal. It is shown that the error in determining the parameters of the current is a few percent. The neural network requires a short time to operate, which potentially allows the use of neural network algorithms in intelligent electronic devices to process current and voltage signals in real time.