Third Zone Distance Relay Enhancement in Power Swing Conditions using IDNN and SMO Algorithm

Abstract

Abstract: In the event that the primary protection fails, remote backup protection is offered using a zone 3 distance relay. However, under demanding conditions like high loads, voltage, and transient instability, the risk of a distance relay malfunctioning is relatively considerable because it jeopardises the stability and dependability of the system. The relay does not run the gearbox line effectively when it is malfunctioning. An advanced power swing blocking strategy has been designed to solve this issue. To prevent distance relay malfunction and boost system dependability, a better DNN-based power swing blocking system is suggested. The improved discrete wavelet transform (IMDWT) is fed sensed data that includes the current and voltage signal of the system. By using the IMDWT, the standard deviation (SD) is computed from the coefficient value of the sensed data, which is used to determine whether a system is in a normal or stressed state. The IDNN's (enhanced Deep Neural Network) most valuable algorithm is given the SD value. The improved DNN in the proposed work functions in two operating modes: RDL-1 (normal condition) and RDL-2 (power swing condition). The threshold-based blocking method improves the performance of the IDNN. The suggested technique determines a suitable system condition based on the threshold value. The results of the suggested method are validated in the Western System Coordinating Council (WSCC) IEEE 9 bus system Software MATLAB/Simulink. The proposed method's total accuracy is 97%. The suggested technique enables quick operation and recognises the power swing state to trip the relay at a distance.