论文部分内容阅读
遗传算法是一种模仿生物自然进化过程的随机搜索和优化算法。其优势在于可以高效地处理传统搜索方法难以解决的非线性问题[1],但简单的遗传算法作为一种启发式搜索算法,寻优理论还不完善,因此在应用中常出现收敛过慢等问题。为了解决这一问题,我们引入H-杂交并提出一种新的改进遗传算法。带H-杂交的改进遗传算法不必要预选交叉概率,变异概率,不必要将目标函数变换成适应度函数,不必要进行变异算子。由于具有这样的特性,带H-杂交的改进遗传算法在一定程度上克服了现有的一些自适应遗传算法的缺陷。
Genetic algorithm is a kind of random search and optimization algorithm that imitates the biological natural evolution process. The advantage of this method is that it can efficiently deal with the nonlinear problems that traditional search methods can not solve [1]. However, the simple genetic algorithm, as a heuristic search algorithm, is not perfect in optimization theory, so it often appears that applications such as slow convergence . In order to solve this problem, we introduce H-hybrid and propose a new improved genetic algorithm. The improved genetic algorithm with H-hybrid does not need to pre-select crossover probability and mutation probability. It is not necessary to transform the objective function into a fitness function. It is unnecessary to carry out mutation operator. Due to such characteristics, the improved genetic algorithm with H-hybrid overcomes the defects of some existing adaptive genetic algorithms to a certain extent.