Study on Improved Fast Immunized Genetic Algorithm

As an effective global optimization method, genetic algorithm has been used in real practice very widely. When it is used in real practice, its slow convergence and poor stability have become the main problems. In order to overcome these problems, from the creation of the initial population, immune selection operation, improved genetic operators, et al, an improved fast immunized genetic algorithm is proposed. Through the simulation experiments of some hard-optimization functions, the proposed algorithm shows its faster convergence and better stability than a lot of existing algorithms´.