Hybrid Approach for Voltage Sag Mitigation in Distribution System

Voltage sag is one of the severe forms of power quality problems in distribution network. It should be mitigated as soon as possible for preventing the load from getting damaged. In the paper, a hybrid approach for voltage sag mitigation in distribution system is proposed. Mitigation is performed by optimally installing DG units in the network. In the hybrid approach, we use two well known Artificial Intelligence techniques (AIs) which are Genetic Algorithm (GA) and artificial neural network (ANN). The proposed approach consists of 3 stage operation. First, a training dataset for ANN with 2 cases is developed using GA. In second stage, ANN is trained and tested with generated training dataset. At end of this stage, ANN is able to provide the optimal locations and sizes for the cases of single DG connection or double DG connection. In third stage, again GA is employed but this time; it is used to optimize the real and reactive powers to be generated by connected DGs. The performance of the proposed approach is tested on IEEE 30 bus system using MATLAB platform. It is showed that the proposed approach is able to mitigate the voltage sag in bus voltages along with the benefits of reducing total power loss and voltage profile improvement in all buses.