You are on page 1of 2

Abstract

As power requests are expanding day by day causing unbalance in the present grid framework which brings about
different causes like load shedding, unbalance voltage and so on which at last influences the end users. Presently to
stay away from every such circumstance the main alternative is to take care of the demand by generation but, world
are additionally slacking with the conventional sources so producing more power is not helpful by traditional ways.

This work focus fulfilling the power requirement by combination of renewable resources.
Combination of renewable resource for any set of power was done by using teacher leaning
based optimization genetic algorithm. Dynamic situation handling capacity of TLBO is basic
feature of this algorithm. In his work initial population contain equal set of solar and wind
power plant in a probable solution. While by two phase learning concept of TLBO population get
optimize by teacher and student section. Here proposed work repeat these steps to get better
solution as per requirement. Experiment was done on real environmental dataset for solar and
wind power power plant. Results were compared with exsisting methods and it was obtained that
proposed work has improved MAE parameter in less execution time while number of power
plant requirement was also reduce.

Index Terms— Electric power Grid, Dynamic load balancing, Genetic algorithm, Hybird Power
Management, Power Loss, Renewable resources.

Fitness Function

For finding difference between two renewable resource Eludician Distance formula was used.
In this work fitness of the probable solution was estimate by total power generation and
average distance of all resources. Here total power T p generation is sum of power generated by
each plant in the probable solution set, while Ad average distance is ratio of summation of each
plant distance from power to smart grid to the number of plants in the solution. So as per
transmission losses rate Lr new power of the probable solution is X.
X = Tp*L*Ad
The Euclidean distance d between two solution X and Y feature vector is calculated by
d = [SUM((X-Y).^2)]^0.5
In above equation Y is required power. This is consider as final rank of the population probable
solution.

Problem Identification

1. Dynamic power requirement handing algorithm is required for smart grid having
renewable power plant.
2. Reduce transmission losses by selecting nearby power plants.
3. Execution time of the model should be less as combination of plant selection be fast.
4. Algorithm should be efficient to find hybid power resources for fulfilling the required
load.

You might also like