摘要
遗传算法是基于生物进化原理的普适性全局优化算法。现定义了交叉相似度的概念，分析了交叉相似度与遗传算法效率的关系，并提出一种基于交叉相似度的自适应遗传算法，其交叉概率、变异概率和编码长度随种群的交叉相似度而变化。理论分析和仿真实验表明改进算法不仅具有以任意精度达到全局最优值的能力，而且具有更高的优化效率。将自适应遗传算法用于丙烯水合反应器的优化，也取得了令人满意的效果。
Genetic Algorithms (GA) are general-purpose global optimization algorithms based on natural evolution principle.In this paper,the concept of crossover similarity is defined,the relationship between the crossover similarity and the efficiency of GA is analyzed.Based on crossover similarity,an adaptive genetic algorithm (AGA) is presented,in which the crossover probability,mutation probability and encoding length can be adjusted to the crossover similarity adaptively.Theoretical analysis and simulation results show that AGA can not only get the global optimum value in arbitrary precision,but also raise efficiency remarkably.The application of AGA in optimal control of a propylene hydration reactor is satisfied.
出处
《石油化工自动化》
CAS
1998年第4期27-30,共4页
Automation in Petro-chemical Industry
关键词
丙烯
水合反应器
遗传算法
优化
Genetic algorithms
Crossover similarity
Adaptive
Optimization
