Transmission line directional protection using neural networks and filtering algorithms

Power transmission lines are the vital links in power systems providing the essential continuity of service from generating plants to the end users. To maintain stability in a power system it is imperative that any fault in the transmission system be identified by protective relays and the faulted line be isolated from the network with minimal delay.; Faults on transmission lines need to be detected, classified and cleared as fast as possible. Moreover, the fault direction should be identified. Development of different modules of a transmission line protective relaying system is outlined in this dissertation. Different modules such as fault direction identification, fault detection and phase selection modules are designed, implemented and tested.; Neural network technique is employed to design the transmission line fault direction identification module. Different neural network structures are used and four different directional modules are proposed.; A new high speed algorithm, based on impedance measurement is proposed for fault detection and phase selection. Adaptive features are also added to the proposed relaying modules to enable them to track the changing operation conditions of the system.; Off-line studies are performed with the proposed relaying modules on a simulated power system model. The system is subjected to different types of disturbances while it is operating at different operating conditions, and the performance of the proposed modules is evaluated. The results obtained indicate that the proposed relaying modules perform rapidly and correctly for different system conditions.; The relaying algorithm has been implemented on a digital signal processor board. Using a power system model consisting of a micro-alternator connected to a constant voltage system extensive experimental studies are conducted and the performance of the relaying algorithm is investigated.; The performance of the proposed modules is investigated further using recorded fault data from a high voltage power system. In this way, the performance of the newly designed relaying modules can be further verified in a more realistic environment. Results using various recorded field data are presented.; The results presented in this dissertation confirm the feasibility of the proposed relaying modules.