动态环境下基于预测机制的多种群进化算法

提出一种动态环境下基于预测机制的多种群进化算法,将预测机制引入到动态进化算法的研究中,对算法所得的某些信息进行记忆,根据记忆序列构建预测模型,当环境发生变化时能够通过预测模型对动态环境进行预先判断.算法采用自组织侦查的多种群策略,多个子种群对搜索子空间进行局部搜索,主种群用于确定新的搜索子空间.在子种群的自适应调整、子种群间的拥挤操作等方面进行了改进,根据子种群所跟踪的最优解位置信息构建预测模型,当环境发生变化时通过预测及子种群的进化实现对动态环境的自适应跟踪.以移动峰问题为测试对象,实验结果表明新算法具有良好的处理动态问题的能力.

Multi-population Evolutionary Algorithm with Forecast Scheme in Dynamic Environment

Forecast scheme is introduced in dynamic optimization algorithms research,and a new multi-population evolutionary algorithm with forecast scheme is proposed.The useful environmental information is stored for building a forecast model,the forecast model is applied to providing prejudges information to algorithm when the environment changes.The algorithm uses multi-population approach named self-organizing scouts;a number of smaller sub-populations are used to track the most promising sub region of search over time,while a larger main-population is continuously global searching for determine new sub search space.In this paper,we discuss principally the adapting adjustment of sub reach spaces,the crowding operation between different sub-populations and the building method of forecast model.The computation results indicate that the new algorithm has the approving performance in dealing the dynamic optimization problems.