基于改进郭涛算法的CCEA函数优化问题  

Improved-GT-Operator-Based Cooperative Co-evolutionary Algorithm to Function Optimization

作　　者：张萍[1] 李涛[1] 李振华[1] 

机构地区：[1]中国地质大学计算机学院,武汉430074

出　　处：《计算机工程》2008年第4期231-232,249,共3页Computer Engineering

摘　　要：郭涛算法在求解函数优化问题方面具有独特的优势,其核心在于多父体杂交。鉴于郭涛算法只有杂交操作而没有变异操作,该文引入高斯正态分布变异算子,提高了对复杂问题的求解效率。分析合作式协同演化算法(CCEA),采用多种群相互作用协同进化的策略求解复杂问题。同时在合作式协同演化模型中引入了郭涛算法,求解复杂高维的函数优化问题。实验结果表明,该模型的效率优于其他模型。Guo Tao（GT） algorithm is highly effective in solving function optimization problems. The core of the algorithm is multi-parent recombination, but it only has crossover operator, no mutation. Gauss mutation operator of Evolution Strategies（ES） is introduced. Cooperative Co-Evolution Algorithm（CCEA） uses multi-population strategy, which means that the fitness of an individual depends on the relationship between that individual and other individuals. CCEA is high-efficient in solving complicated problem. A cooperative co-evolution model to function optimization is proposed, based on an improved Guo Tao operator, which employs a Gauss mutation operator to enhance its exploring ability. Experimental results show that the model is more effective than others.

关 键 词：郭涛算法 高斯变异算子 合作式协同演化算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]