Generating initial population of farmland fertility algorithm with chaotic maps

Optimization is defined as choosing the best solution among all possible solutions of a problem under certain conditions. Classical methods used to solve optimization problems may be insufficient to solve real life problems. In addition, proportional to the size of the problem, the solution times and the complexity of the solution is increasing. These disadvantages of classical methods have led to the development of metaheuristic research methods. Parallel to the development of computers, metaheuristic research methods that have been studied for the last four decades are inspired by nature. It is the subject of scientists' researches that living species have an optimum design in the ecosystem and find the optimum solution in their behavior and foraging behaviors, the physical process and laws of natüre ,contributes to the increase of nature-inspired algorithms. In this thesis, the recently proposed Farmland Fertility Algorithm is discussed in detail. Chaotic maps were used to generate the initial population of the Farmland Fertility Algorithm. Initial populations with chaotic maps were formed by 3 different methods. 10 chaotic maps were used in each method. Each chaotic map algorithm was tested with 8 quality test functions. As a result of the tests, it was observed that the results of the approach of Generating the Whole Population and all dimensions with chaotic maps and the third approach in which the population was produced with one dimensional chaotic maps were generally similar or better with the Farmland Fertility Algorithm.In the second approach, it has been observed that the Farmland Fertility Algorithm, in which the tent (Tent) chaotic map is applied, produces more successful results for almost all dimensions and iterations. In the third approach, successful results were obtained in most of the chaotic map algorithms.