An intelligent wire fault diagnosis approach using time domain reflectometry and pattern recognition network

The aim of this paper is to develop a new approach for wiring networks diagnosis. This approach is based on time domain reflectometry (TDR) and pattern recognition network all combined with a simple localisation and characterisation algorithm. The developed approach has the advantage of being fast, reliable and accurate. The TDR responses are generated using the analytical transmission line method. Furthermore, the manipulated data are reduced since only the main peaks are exploited. The performance of the proposed approach has been assessed using four cases, two using a simple Y-shaped network and two using a more complicated YY-shaped network. The obtained results are very satisfactory, and they show the applicability of the proposed approach for real systems.