Abstract

As the electricity demand increases dramatically in Sweden, the need of using the existing electricity grid as efficiently as possible gains more importance. Simultaneously as needs expand, so does production in the form of wind parks and solar parks. This has led to an increase in connection requests at Svenska Kraftnät, the Swedish transmission system operator. The current process for accepting or rejecting these requests is based on a first-come-first-serve principle, where each request is investigated separately. This thesis investigates an alternative way of processing the requests in clusters and optimizing which combination of them is the best one to accept from a technical point of view. To handle this multiobjective combinatorial optimization problem, a multiobjective Genetic Algorithm with a Pareto filter is developed. The Genetic Algorithm finds a refined Pareto front containing optimal solutions that are plotted with objective function values. The user can then easily analyze the optimal solutions and decide upon which the final optimal request combination is. The developed Genetic Algorithm reaches a close-optimal Pareto front estimation after exploring between 15-40% of the solution space.