Skip to Main Content
Several neural network techniques have been proposed to solve the problem of automatic fault diagnosis in power systems. In this paper. A new and accurate parallel neural network-based approach for fast fault diagnosis and alarm handling of power systems is developed which overcomes some of the limitations of existing techniques. Three different neural-network configurations are proposed and their performances are compared using the data of an actual high-voltage power network.
Date of Conference: 30 Oct.-1 Nov. 2000