An adaptive genetic algorithm

According to the porblems of the simple genetic algorithm′s weak stable property,the premature convergene and easily getting into local optimum,a new adaptive genetic algorithm is presented which is based on the crossover probability and mutation probability.In order to enlarge the search space rapidly and keep the variety of population at a stable level,through changing crossover probability and mutation probability automatically with fitness,the crossover and mutation operation are used on different individual purposefully,The simulation experiments show that this algorithm has great advantage of convergence property over simple genetic algorithm,and it can effectively avoid the premature convergence problem caused by the high selective pressure in simple genetic algorithm.Moreover,the algorithm improves the ability of searching an optimum solution and increases the convergent speed.