A hybrid method for optimal load shedding and improving voltage stability

In this paper, a hybrid method is proposed for reducing the amount of load shedding and voltage collapse. The hybrid method is the combination of Genetic Algorithm (GA) and Neural Network (NN). The GA is used by two stages, one is to frame the optimization model and other stage is to generate data set for developing the NN based intelligent load shedding model. The appropriate buses for load shedding are selected based on the sensitivity of minimum eigenvalue of load flow Jacobian with respect to the load shed. The proposed method is implemented in MATLAB working platform and the performance is tested with 6 bus and IEEE 14 bus bench mark system. The result of the proposed hybrid method is compared with the GA based optimization algorithm. The comparison shows that, the proposed method ensures voltage stability with minimum loading shedding.