Intelligent phase comparison carrier protection

Conventional phase comparison carrier protection mainly consists of three units-starting unit, phase discriminator and trip relay. The phase discriminator is the brain of phase comparison protection. It compares the measured phase shift of the high-frequency carrier current signals θ at the two ends of the protected transmission line with the stability angle φ, to determine whether or not to trip the line. It is known that θ varies under different power system operating conditions and fault conditions. Since in conventional phase comparison protection φ is set off-line and cannot change on-line to adapt to various operating and fault conditions, the possibility of maloperation exists. In order to overcome the shortcoming we have developed an artificial neural network (ANN) approach for phase comparison carrier protection. There are some advantages which are inherent in ANN such as excellent noise immunity, robustness, fault-tolerance and generalization capability. A multilayer feedforward network (MFFN) approach for phase comparison protection is presented in this paper. Such an approach has excellent generalization capability. Digital simulation has been carried out using the Electromagnetic Transient Program to verify the performance of the proposed approach. Results show that the approach has the ability to adapt to various operating and fault conditions