This paper proposes a new prognostic method capable of preventing catastrophic failures in Medium Voltage (MV) power cables. The main objective is the development of a monitoring system focused on the detection and localization of cable overtemperatures in underground distribution networks. The predictive analysis proposed here is based on a Multi-Layer neural network with Multi-Valued Neurons (MLMVN), which elaborates measurements of high frequency signals transmitted through Power Line Communication (PLC) devices. Therefore, the prognostic method does not require the introduction of additional components since the power line is already equipped with a communication system. This allows low intrusion and the possibility of monitoring power lines during their operation. Furthermore, the MLMVN-based classifier processes magnitude and phase of the received signals without preliminary coding steps and ensures a low computational cost. The main theoretical concept on which the predictive analysis is based is the detection of malfunctions starting from their effects on the cable parameters. For this reason, an RG7H1M1 cable has been experimentally characterized in the frequency range between 90 kHz and 1 MHz, both in nominal conditions and in overheating situations generated by means of a climatic chamber. The changes in the electrical parameters of the cable modify the transmitted signal and the monitoring system proposed here allows the identification and localization of the overheated section with high accuracy.

Frequency Characterization of Medium Voltage Cables for Fault Prevention through Multi-Valued Neural Networks and Power Line Communication Technologies / Bindi M.; Luchetta A.; Lozito G.; Carlo CAROBBI ; Grasso F.; Piccirilli M.. - In: IEEE TRANSACTIONS ON POWER DELIVERY. - ISSN 0885-8977. - ELETTRONICO. - (2023), pp. 1-11. [10.1109/TPWRD.2023.3270128]

Frequency Characterization of Medium Voltage Cables for Fault Prevention through Multi-Valued Neural Networks and Power Line Communication Technologies

Bindi M.;Luchetta A.;Lozito G.;Carlo CAROBBI;Grasso F.;Piccirilli M.
2023

Abstract

This paper proposes a new prognostic method capable of preventing catastrophic failures in Medium Voltage (MV) power cables. The main objective is the development of a monitoring system focused on the detection and localization of cable overtemperatures in underground distribution networks. The predictive analysis proposed here is based on a Multi-Layer neural network with Multi-Valued Neurons (MLMVN), which elaborates measurements of high frequency signals transmitted through Power Line Communication (PLC) devices. Therefore, the prognostic method does not require the introduction of additional components since the power line is already equipped with a communication system. This allows low intrusion and the possibility of monitoring power lines during their operation. Furthermore, the MLMVN-based classifier processes magnitude and phase of the received signals without preliminary coding steps and ensures a low computational cost. The main theoretical concept on which the predictive analysis is based is the detection of malfunctions starting from their effects on the cable parameters. For this reason, an RG7H1M1 cable has been experimentally characterized in the frequency range between 90 kHz and 1 MHz, both in nominal conditions and in overheating situations generated by means of a climatic chamber. The changes in the electrical parameters of the cable modify the transmitted signal and the monitoring system proposed here allows the identification and localization of the overheated section with high accuracy.
2023
1
11
Bindi M.; Luchetta A.; Lozito G.; Carlo CAROBBI ; Grasso F.; Piccirilli M.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1312514
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact