中文摘要：

连续域分布估计算法一般假设数据服从高斯分布,而且大多采用了单峰的概率模型,但是对于一些复杂的优化问题,单峰的高斯分布模型不能有效地描述解在空间的分布.本文提出一种基于序贯重点采样粒子滤波的分布估计算法,采用带权粒子描述优选集样本服从的概率分布并从中采样得到下一代种群,不需要假设样本服从高斯分布,并且算法采用的概率模型是多峰的.仿真实验结果验证了本文方法的正确性和有效性.

英文摘要：

Estimation of distribution algorithm in continuous domains is generally based on assumption that variables subject to Gaussian distribution and that the probability model is single-peaked one,which is not capable of describing the solutions distribution effectively for complex optimization problems.Aiming to improve such drawback,an estimation of distribution algorithm depending upon sequential importance sampling particle filters is presented.In this algorithm,the variables are not required to subject to Gaussian distribution.Instead,the distribution of samples is represented by weighted particles and the used probability model is multi-peaked.The next generation of population is produced from the above distribution.The simulation results indicate the validity of the algorithm.