NEURAL NETWORKS FOR FAULT LOCATION IN SUBSTATIONS

Faults producing load disconnections or emergency situations have to b e located as soon as possible to start the electric network reconfigur ation, restoring normal energy supply. This paper proposes the use of artificial neural networks (ANNs), of the associative memory type, to solve the fault location problem. The main idea is to store measuremen t sets representing the normal behavior of the protection system, cons idering the basic substation topology only, into associative memories. Afterwards, these memories are employed on-line for fault location us ing the protection system equipment status. The associative memories w ork correctly even in case of malfunction of the protection system and different pre-fault configurations. Although the ANNs are trained wit h single contingencies only, their generalization capability allows a good performance for multiple contingencies. The resultant fault locat ion system is in operation at the 500 kV gas-insulated substation of t he Itaipu system.