摘要
多目标优化问题是目前遗传算法应用研究的一个重点。本文针对经典遗传算法在多目标优化计算中,难以获得足够的比较均匀的Pareto优集的不足,提出一种热力学遗传算法,研究热力学中熵和温度的概念,并综合利用约束交叉、适应度共享技术来进行多目标函数的优化计算。实验结果显示,这种改进型遗传算法能得到一个较好的Pareto优集。
Multiobjective optimization is one of the main research fields of genetic algorithm, hut it is hard to get adequate and well - distributed optimal solutions when traditional genetic algorithm is used. In this paper, we propose a thermodynamical genetic algorithm in which the concepts of temperature and entropy in thermodynamics are used and are combined with the Pareto - based ranking and fitness sharing. The experiment results show that this algorithm can find a better of Pareto optimal solutions.
出处
《计算机应用与软件》
CSCD
2000年第11期19-23,54,共6页
Computer Applications and Software
基金
国家自然科学基金资助项目(编号:69671022)
关键词
多目标优化
解
遗传算法
模拟退火算法
Genetic algorithm Multiobjective optimization Pareto optimal solutions Entropy Thermodynamical genetic algorithm
