As a global optimization algorithm, genetic algorithm has been widely used in many areas. However, the classical genetic algorithm has some problems, such as low convergence speed, premature convergence easily and so on. Some of the existing adaptive genetic algorithm often falls into a local optimal solution, and the stability is very poor. So in this paper, we have a further study on the adaptive genetic algorithm in order to improve the algorithm performance. In this paper, we adopt new selection strategy, propose two calculating methods of crossover probability and a new continuous mutation probability calculating method, in addition, the grading thinking is used in this paper. Numerical experiments show that new algorithms have good performance in searching global optimal algorithm. The decision-making method with a new algorithm on power purchase and distribution planning can obtain an excellent distribution scheme for the example which proposed in the paper. |