In this paper we propose an analysis of the possibilities of spikes detection which appears on the power lines of electrical distribution. The importance of this problem consists in the possibility of surveillance at distance the regimes of power electrical motors. This regime was studied in an industrial network which supplies the power pumps necessary to assures a city water consumption. We consider the industrial electrical networks as a linear, time invariant and causal system which has a certain response to the electrical spikes, assimilates with Dirac impulses. For that reason the output signal represents the impulse response of the system. The main problem is to detect this signal and to establish the moments of its presence. To detect and extract the useful signal linear methods like fourier transform or Short Time Fourier Analysis are not very relevant. A more efficient method to detect the spikes presence consists in using quadratic detectors in timefrequency domain, like detectors based on Wigner-Ville transform or detectors based on Ambiguity Function. In this paper we tried to present the advantages of these last detectors.

Date Published: 14 December 2016
PDF: 9 pages
Proc. SPIE 10010, Advanced Topics in Optoelectronics, Microelectronics, and Nanotechnologies VIII, 100102A (14 December 2016); doi: 10.1117/12.2243198

Published in SPIE Proceedings Vol. 10010:
Advanced Topics in Optoelectronics, Microelectronics, and Nanotechnologies VIII
Marian Vladescu; Cornel T. Panait; Razvan Tamas; George Caruntu; Ionica Cristea, Editor(s)