Microgrid energy management using metaheuristic optimization algorithms

Abstract

This article addresses the economic dispatch problem of microgrids. Firstly, it presents the application of both traditional and newly introduced metaheuristic optimization algorithms to solve for the optimal power flow problem for the IEEE 30 bus system after which the best performing algorithm is chosen for cost-effective economic dispatch in a microgrid designed upon the microgrid facility present at Wroclaw University of Science and Technology. All algorithms investigated have been combined with the academic power analysis tool, MATPOWER. The idea behind the approach is to find a compromise between the solution search capabilities of the metaheuristics and the optimized performance of MATPOWER. The algorithms explored include 3 traditional algorithms which are the genetic algorithm, particle swarm optimization and mixed integer distributed ant colony optimization and 2 recently developed algorithms which are the political optimizer and the Lichtenberg algorithm. Hyperparameter tuning was carried out for all investigated algorithms. The results have shown that the ant-colony based algorithm is the most suitable of all the choices in terms of having the best convergence time of 19.17 s, a final solution value of 801.57 ($/h) and reliability in terms of reproducing the best solution for the test system. It is then used for economic dispatch which is guided by an objective function that minimizes the levelized cost of energy in the microgrid.