A novel grading scheme for loads to optimize load shedding using genetic algorithm in a Smart Grid environment

In developing countries, a large growth in power demand necessitates the efficient distribution of available power. Whenever power demand is more than the power generation load shedding is carried out. Under the current scheme, load shedding is done by disconnecting an entire feeder, employing `Round robin´ technique. In most cases, this method fails to shed exact amount of load resulting in either over-shedding or under-shedding. Further, load shedding is done regardless of the type of the load connected to a feeder. This Paper proposes a novel grading scheme for loads to minimize the impact of load shedding by taking revenue loss and social factors into consideration. The genetic algorithm developed minimizes the error between the amount of load to be shed and the actual load shed, simultaneously optimizing the overall impact of load shedding. The algorithm is developed for a smart grid environment, assuming a two way communication between the consumer and the utility is present. The algorithm is tested on a sample system comprising practical feeder data.