【摘要】 提出了一种用于多目标优化的多概率模型分布估计算法,该算法在进化的每一代中使用多个概率模型来引导多目标优化问题柏拉图(Pareto)最优域的搜索。分布估计算法使用概率模型引导算法最优解的搜索,而使用多个概率模型可以保持所得多目标优化问题最优解集的多样性。该算法具有很强的寻优能力,所得结果可以很好地覆盖Pareto前沿。实验通过优化一组测试函数来评价该算法的性能,并与其它多目标优化算法进行了比较,结果表明该算法相比于其它同类算法可以更好地解决多目标优化问题。更多还原

【Abstract】 A new Estimation of Distribution Algorithm based on multi-probability model for multi-objective optimization is presented. This algorithm guides searching Pareto-front of multi-objective optimization problem by using multi-probability model at each generation. Estimation of Distribution Algorithms use probabilistic model to search for problem’s optimal solutions, and using multi- probability model can maintain the diversity of multi-objective optimization problem’s optimal set. This algorithm has the powerful ability of searching optimal, and the result can cover the Pareto-front. A set of experiments has been implemented to evaluate the performance of this algorithm by optimizing a group of test function set, and compare with other multi-objective optimization algorithms. The results show that the new algorithm presented can perform better in solving multi-objective optimization problems.更多还原