A Scalable Memetic Algorithm For High Dimensional Power System Optimization Problems According To Systematic Mapping On Software Engineering To Achieve More Performance And Productivity
The modern power system optimization problems have been facing increasing challenges such as the existence of big number of objective functions and decision variables. Multi-objective Evolutionary Algorithms (MOEAs) have been widely used to solve such complex real-world power system applications that have more than one objective function. However, most traditional MOEAs work well when the number of objective functions is less than three. The performance of MOEAs starts degrading significantly when the number of objective functions increases. This work investigates the application of new scalable evolutionary algorithm to solve optimization problems of power systems that have many objective functions. Specifically, we propose a new evolutionary algorithm based on the NSGA-II algorithm to efficiently solve the many-objective optimization problems. The proposed algorithm merges three efficient schemes which are: a new efficient sorting method, smart archive and simple local search to speed up the solutions convergence process to the POF and enhance the diversity of the solutions. A set of multi-objective test problems are presented in this research to show the performance and applications of the proposed optimization algorithms. The results show that our proposed algorithm significantly outperforms the other algorithms when the number of objective functions is high.