论文部分内容阅读
针对如何通过附加的方法对多目标化问题进行理论分析,提出并证明了选择附加函数的3个前提条件.提出一种多目标化进化算法,根据种群中个体的多样性度量进行多目标化,并采用改进的非劣分类遗传算法对构造所得的多目标优化问题进行多目标优化.在静态和动态两种环境下进行算法性能验证,结果表明,在种群多样性保持、处理欺骗问题、动态环境下的适应能力等方面,所提算法明显优于其他同类算法.
Aiming at how to analyze the multi-objective problem through additional methods, three preconditions for selecting additional functions are proposed and proved.A multi-objective evolutionary algorithm is proposed based on the multi-objective measure of individual diversity in the population, And the improved non-inferior genetic algorithm is used to optimize the multi-objective optimization problem.Analyzing the performance of the algorithm under both static and dynamic conditions, the results show that in the diversity of the population, the handling of fraud, dynamic environment Adaptability and other aspects, the proposed algorithm is significantly better than other similar algorithms.