Abstract—In the coming years, the flexibility of distribution network will be required more due to the implementation of distributed generation of renewable energy sources such as wind, solar, biomass, etc. The distributed generations are connected to the power grid in order to contribute to supplying energy for the loads. Thus the power flow direction and fault direction at the relays will be also changed. Therefore, using directional overcurrent protection relay is necessary to isolate the fault quickly and exactly. This paper proposes a novel method based on linear adaptive neural network which is used in directional overcurrent protection relay. The proposed method has the advantages compared with the fast Fourier transform method in signal processing. It can estimate the amplitude and phase angle of voltage and current signals accuracy after each updated sample. In addition, the proposed method is verified on the distribution network with/without distributed generation which is modeled on Matlab/Simulink software. The simulation results show that the proposed method accurately detect the direction at the relay when the fault occurs at different locations on the network. 
 

Index Terms—ADALINE, directional protection relay, distributed generator, smart grid

Cite: Doan Duc Tung and Ngo Minh Khoa, "A Novel Algorithm of Directional Overcurrent Protection Relay Based on Adaptive Linear Neural Network," International Journal of Electronics and Electrical Engineering, Vol. 4, No. 6, pp. 494-499, December 2016. doi: 10.18178/ijeee.4.6.494-499