基于分布估计算法的输电网扩展规划

TRANSMISSION NETWORK EXPANSION PLANNING BASED ON ESTIMATION OF DISTRIBUTION ALGORITHMS

分布估计算法是一类新的进化算法,它通过统计在当前群体中选出的个体信息给出下一代个体分布的概率估计,用随机取样的方法生成下一代群体。文章将分布估计算法应用于求解输电网扩展规划问题,提出了两种基于分布估计算法(基于群体的递增学习算法和因子分布算法)的电力系统输电网扩展规划模型,分析了加权估计、随机母本规模选择、条件概率链的重新排列、随机变异和精英保留等改进策略对算法的影响。仿真分析结果表明了文中所采用的分布估计算法在求解输电网扩展规划问题时是可靠有效的。

Estimation of distribution algorithms (EDA) is a new kind of evolution algorithm, through the statistics of the information of selected individual in current group the probability estimation of the individual distribution in next generation is given, and the next generation of group is formed by random sampling. Here, the EDA is applied to solve the transmission network expansion planning problem, two models of transmission network expansion planning based on EDA, i.e., the population-based incremental learning: (PBIL) and factorized distribution algorithm (EDA), are put forward. The influence of several strategies, such as weighted estimation, random size of parents set selection, readjusting of the conditional probability sequence, stochastic mutation and elitist reserved on the algorithm is analyzed. Simulation results show that the used EDA is reliable and effective for solving the transmission network expansion planning problems.