Development of micro-artificial algae algorithm for solution of optimization problems

In recent years, metaheuristic algorithms have been frequently used to solve optimization problems. These solution methods do not provide sufficient time and cost benefit in complex and high parameter optimization problems. However, due to the lighter hardware requirements and the possibility of working in embedded systems with memory saving approach, Micro-Metaheuristic Algorithm methods have been proposed in the literature. Many Micro-Metaheuristic Algorithms have been introduced by the researchers in the form of microstructures that enable them to produce solutions at a lower cost by developing solutions to accelerate problem solving such as creating and re-updating small-sized populations and protection of individuals. In this thesis project, a study has been carried out on a micro artificial algae algorithm with high performance in solving optimization problems by using low population on the artificial algae algorithm and adding auxiliary methods without disturbing the algorithm's working structure. The study was compared with Micro Particle Swarm Optimization (μPSO), Micro Bacteria Nutrient Foraging Optimization Algorithm (μBFOA) and standard Artificial Algae Algorithm (AAA) and it was observed that it produced successful results. Apart from the functions using the algorithms compared with the proposed method, it was observed that CEC2015 functions were stuck in local areas in experiments with CEC2015 functions and research and studies were carried out to overcome this problem. With these procedures, the main proposed method, Micro Artificial Algorithm with Levy (µAAAlevy) was presented. This algorithm has shown to be a promising algorithm compared to AAA with 3 and 40 populations.