摘要
通过对多目标优化方法研究现状的分析,针对多目标优化问题的特点提出一种基于联合正态分布的求解多目标优化问题的分布估计算法。该算法将不同目标函数的自变量作为不同的随机变量,用联合正态分布对扩充的解空间进行估计,并用投影的方式将采样点映射到解空间中。仿真结果表明了该算法的有效性。
By analyzing the research status of Multi-Objective optimization, an Estimation of Distribution Algorithm for Multi-Objective Problem based on joint normal distribution is proposed. In this algorithm, variables of different objective functions are regarded as different random variables, and the extended solution space is estimated based on joint normal distribution, and the samples are mapped to the solution space by projection. Simulations prove the effectiveness of the algorithm.
出处
《太原科技大学学报》
2010年第1期55-58,共4页
Journal of Taiyuan University of Science and Technology
基金
太原科技大学青年基金(2007116)
关键词
分布估计算法
多目标优化
正态分布
estimation of distribution algorithm, multi-objective optimization, normal distribution
