粒子群算法的不确定动态多目标优化方法研究

发布时间：2018-04-02 09:36

本文选题：粒子群算法　切入点：不确定　出处：《科学技术与工程》2017年15期

【摘要】：当前不确定动态多目标优化方法通常将多目标问题转换成单目标问题,将其他目标看作约束条件,仅可得到单个解,无法有效体现不确定多目标之间的关系,导致得到的解质量低。为此,提出一种新的基于粒子群算法的不确定动态多目标优化方法,给出不确定动态多目标优化问题的数学描述,介绍了粒子群算法,针对粒子群算法容易陷入局部最优的弊端,引入动态变异算子对其进行改进,通过改进的位置更新公式实现粒子群算法位置的自适应更新,给出解决不确定多目标优化问题的详细过程,在此基础上,通过分段线性函数参数化实现不确定动态多目标优化。实验结果表明,所提方法搜索能力强,采用所提方法得到的解与真实解最相近,质量最高。
[Abstract]:At present, the uncertain dynamic multi-objective optimization method usually transforms the multi-objective problem into a single-objective problem. The other targets are regarded as constraints, and only a single solution can be obtained, which can not effectively reflect the relationship between the uncertain multi-objective. Therefore, a new uncertain dynamic multi-objective optimization method based on particle swarm optimization is proposed. The mathematical description of the uncertain dynamic multi-objective optimization problem is given, and the particle swarm optimization algorithm is introduced. In order to solve the problem that particle swarm optimization (PSO) is easy to fall into local optimum, dynamic mutation operator is introduced to improve PSO, and the adaptive updating of PSO's position is realized by the improved location updating formula. The detailed process of solving the uncertain multi-objective optimization problem is given. On the basis of this, the uncertain dynamic multi-objective optimization is realized by piecewise linear function parameterization. The experimental results show that the proposed method has strong searching ability. The solution obtained by the proposed method is most close to the real solution and has the highest quality.
【作者单位】：中原工学院;
【基金】：国家自然科学基金(51475290、51075261) 2015年度河南省软科学研究计划项目(152400410598)资助
【分类号】：TP18

【相似文献】