Autors: Mateev, V. M., Ralchev, M. L., Marinova, I. Y.
Title: Electric Arc Discharge Power Estimation by CNN Image Classification
Keywords: CNN; Convolutional neural network; Electrical arc discharge; FFT harmonic analysis; Image processing

Abstract: In this work, a machine learning approach for the estimation of electric arc discharge parameters is presented. The machine learning approach is implemented by a multilayer convolutional neural network (CNN) that is trained over sequential images of a video stream of an electric arc discharge, recorded in the visible light spectrum. Image datasets, which are obtained by actual discharge observations, are combined with transient electric parameter values, such as electric current, voltage and power, obtained by simultaneous measurements. A CNN architecture is evaluated to predict the electric active power on such visual images.

References

    Issue

    Lecture Notes in Electrical Engineering, vol. 886, pp. 315 - 326, 2022, Switzerland, Springer Nature, https://doi.org/10.1007/978-3-030-98886-9_25

    Copyright Springer International Publishing

    Вид: публикация в международен форум, публикация в реферирано издание, индексирана в Scopus