Open AccessProximity effect modelling for cables of finite length using the hybrid partial element equivalent circuit and artificial neural network method
Author(s) -
Chen Hongcai,
Du Yaping
Publication year - 2018
Publication title -
iet generation, transmission and distribution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.92
H-Index - 110
eISSN - 1751-8695
pISSN - 1751-8687
DOI - 10.1049/iet-gtd.2018.5392
Subject(s) - artificial neural network , finite element method , computer science , equivalent circuit , partial element equivalent circuit , element (criminal law) , biological system , engineering , artificial intelligence , structural engineering , voltage , electrical engineering , biology , law , political science
This study presents an efficient method for modelling the proximity effect in complex conductor systems. This method is based on a discretisation partial element equivalent circuit (DPEEC) scheme in combination with artificial neural network (ANN). Circuit parameters of a conductor system are obtained with DPEEC at low frequency. ANN trained with the low‐frequency parameters is employed to predict proximity effect at high frequencies. The proposed method significantly improves the calculation efficiency in both time and memory consuming. The method is validated by comparing with the result obtained by MoM‐SO. Case studies of closely‐spaced cables with different configurations are analysed. It is applied to evaluate the lightning current in typical cable installations. The comparison among different configurations reveals that the proximity effect leads to uneven current distribution in cables. Cable modelling without considering the proximity effect could lead to significant errors in transient current analysis.