摘要
遗传算法是一类模拟自然界生物进化过程与机制、求解问题的自组织和自适应的人工智能技术,是非常好的求解优化问题的算法,但是它也容易产生早熟现象,且局部搜索能力较差。因此,在分析传统的非线性规划方法的基础上,针对传统方法的局限性,为非线性规划模型设计了一种新的启发式算法,即结合遗传算法、模拟退火算法和动态惩罚函数法的混合遗传算法,以发挥各算法处理问题的优势。对算法的过程进行了分析。通过实例说明,该算法对于求解所建立的问题是有效的。
Genetic algorithm is a kind of self-organizing and self-adaptive intelligent technology to solve the problem by simulating the evolution process and mechanism of living beings in the natural world. It is an effective algorithm to solve the optimization problem. However, a genetic algorithm tends to be trapped in prematurity, and often has a poor local searching ability. So in this paper, based on analytic methods of the nonlinear programming and to solve the limitations of these methods, a new elicitation algorithm to solve the nonlinear programming problem-hybrid genetic algorithm with the advantages of genetic algorithm, simulated annealing algorithm and dynamic penalty function algorithm, is presented. The process of the hybrid genetic algorithm is analyzed, an illustration is given to show that the algorithm is effective.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2003年第5期621-624,共4页
Systems Engineering and Electronics
基金
南开大学天津大学刘徽应用数学中心资助课题
关键词
惩罚函数
分类约束
混合遗传算法
Penalty function
Class constraints
Hybrid genetic algorithm
