| Genetic algorithm is a kind of search algorithm based on natural genetics and natural selection mechanism; it uses the important mechanism of the natural biological system for designing the artificial system. Genetic algorithm has the advantages of simple and global search, which makes the genetic algorithm widely used in the field of system control, numerical optimization and other fields. However, the classical genetic algorithm is prone to premature convergence and fall into the local optimal solution easily, so it has great significance to improve classical genetic algorithm.Firstly, we briefly introduce the content of genetic algorithm. For the problem of premature convergence of classical genetic algorithm, the crossover probability and mutation probability formula are improved. The number of the best individual unchanged during evolution process is introduced into the formula of crossover and mutation probability, the ability of genetic algorithm to jump out of local optimal solution is enhanced by adjusting the number of generation and the number of the best individual unchanged. To a certain extent, it can improve the search performance of classical genetic algorithm.Genetic algorithm is used to solve nonlinear programming problems. Some infeasible solutions which have high fitness and near feasible domain boundary are preserved in the population by introducing the quasi simplex operator. Thus, genetic algorithm can search the two directions from the feasible region and feasible region. It can increase the diversity of population genetic, which makes the search process more effectively.The active power and reactive power dispatch optimization model of the wind farm which is Composed by double feedback induction generator is given. The problem is a kind of nonlinear programming problem. The improved genetic algorithm is used to solve the optimization model. The better scheduling scheme is obtained. To a certain extent, it can reduce the loss and has good application value. |
PDF Full Text Request