关键词: 遗传算法, 梯度信息, 指导交叉, 最速下降法, 最优解

Abstract: Concerning the blind search for cross-individual in the solution space, which leads to low efficiency and low convergence speed in the later stage of simple Genetic Algorithm (GA), an improved instructed-crossover genetic algorithm based on gradient information was proposed. It executed crossover operation via choosing special individuals from the set range of the negative gradient of the objective individual got from the current population, and made the offspring more closer to the optimal solution. It guaranteed the purpose and feasibility of the crossover operation. The simulations on four typical test functions indicate that the proposed algorithm can greatly improve the efficiency and precision in searching the optimum value.

Key words: Genetic Algorithm (GA), gradient information, instructed crossover, steepest descent method, optimum value