论文部分内容阅读
为求解复杂函数优化问题,基于人类记忆原理和人际关系,提出了一种新型函数优化方法,即MP-IRO算法。在该算法中,将个体分为恋人、知己、敌人、小人、陌生人5种对象类型,对应于恋爱、聚集、攻击、排斥、防御等5种行为,并构造相应的演化算子。恋爱算子能优先选择拥有长时记忆的个体,分享其表征特性;聚集算子能使个体摆脱局部最优解的陷阱;攻击算子能使个体之间活跃度增强;排斥算子能让个体远离瞬时记忆试探解方向,扩大搜索范围;防御算子能增加随机性。测试结果表明,本算法对求解复杂函数优化问题具有较高的适应性和收敛速度。
To solve the problem of optimization of complex functions, a new method of function optimization, MP-IRO algorithm, is proposed based on human memory theory and human relations. In this algorithm, we classify individuals into five types of objects: lover, confidant, enemy, villain and stranger, corresponding to five kinds of behaviors such as love, aggression, aggression, exclusion and defense, and construct corresponding evolution operators. Love operator can give priority to individuals who have long-term memory and share their characterization; Aggregation operator can make individuals get rid of the trap of local optimal solution; Attack operator can enhance the activity between individuals; Exclusion operator allows individuals Away from the instantaneous memory test solution direction, to expand the search; defense operator can increase the randomness. The test results show that the proposed algorithm has high adaptability and convergence speed for solving complex function optimization problems.