摘要
为找到一种快速有效的配电网重构算法,提出了一种结合贝叶斯统计推断理论和分布估计思想的混合算法,利用这种算法进行配电网络的优化调整。通过贝叶斯推断理论建立分布估计算法的概率模型,用先验分布和条件分布模型产生后验概率模型,用后验概率代替先验概率采样产生新个体组成优势群体,再用优势群体中的个体信息更新各概率模型,用更新后的后验概率不断采样直到寻到最优解。最后,通过和单纯分布估计算法及免疫二进制粒子群算法的重构结果相比较证明了这种混合算法的快速、有效。
In order to seek a rapid and effective algorithm of distribution network reconfiguration,a hybrid algorithm combined with Bayesian statistical-inference and distribution estimation is proposed.Via the algorithm for optimization and adjustment of distribution network,the probability distribution model through the theory of Bayesian inference is established,then a posteriori probability model with the prior distribution and the conditional distribution is produced,and new individuals are generated into advantage of group with the posteriori probability of sampling,updating these probability model by advantage of group.It will be sampled continuously with the updated posterior probability until the optimal solution can be otained.Finally,the comparative result with estimation of distribution and immune binary particle swarm proves that the hybrid algorithm is rapid and effective.
出处
《电力系统及其自动化学报》
CSCD
北大核心
2014年第6期54-59,共6页
Proceedings of the CSU-EPSA
关键词
分布估计
贝叶斯推理
配电网重构
后验概率
免疫二进制粒子群
estimation of distribution
Bayesian inference
distribution network reconfiguration
posterior probability
immune binary particle swarm
