Research Of Improvement And Application Of A Kind Of Optimization Algorithm

In order to meet the development of both military and industrial production, some optimization problems were proposed in the 1930 s. However, these problems could not be solved by using both the classical differential method and the classical variational method. Under the concerted research of many scholars, optimization method arised at the historic moment. It is a kind of new mathematical method and has been widely used because of its important practical value in recent decades.Genetic algorithm has a history of nearly forty years as a kind of heuristic optimization method and is widely used in major areas with its unique advantages. But it also has some limitations as a kind of optimization method. Therefore, this paper studied the following contents.The first chapter mainly introduced the research background, the development course, previous research results in recent decades and the applications of genetic algorithm and the main research contents of this paper were given.The second chapter mainly introduced some preliminary knowledge in this paper, such as definition, terminology and basic principles. And the principles include coding, fitness function, genetic operation, parameter selection and algorithm design.The third chapter listed several common methods in the present to solve the optimization problem including several classical algorithms and several heuristic algorithms, and expounded the advantages and disadvantages of them.The fourth chapter mainly studied the premature convergence problem of genetic algorithm, and improved the crossover probability cp and mutation probability mp which are parameters based on the IAGA adaptive genetic algorithm. Then, the new general expressions of cp and mp were given. Finally, the numerical results showed the feasibility and effectiveness of the improved algorithm.