基于改进微分进化算法的电力系统无功优化

Reactive power optimization in power system based on modified differential evolution algorithm

针对传统无功优化模型对于大系统编程实现十分困难的缺点,采用矩阵形式的数学模型进行优化。对标准微分进化算法提出了改进,引入增强算子,并采用模拟赌盘操作的方法有目的地使种群中的较差个体参与增强运算,提高了算法的寻优能力。由于微分进化算法中,初始种群的优劣对算法的收敛性有重要影响,根据无功优化的本质,依据初始潮流结果启发初始种群产生,加快收敛速度。在IEEE-14系统上进行校验,并与其他方法比较,结果表明,提出的改进算法具有收敛特性好、运行速度快的突出优点。

In order to overcome the programming difficulty of traditional reactive power optimization model for large-scale systems,the model of reactive power optimization in the form of matrix is presented.The modified DE algorithm is improved by introducing enhanced operator and operating analog gambling site by purpose to make the poor individuals of population participate in enhancing operation and improving the method of optimizing capacity.The merits and demerits of initial population have a significant impact on the algorithm convergence.According to the nature of reactive power optimization,initial population based on results of the power flow.MPDE applied for optimal reactive power is evaluated on an IEEE-14 bus power system.It is shown that the proposed approach has better convergence and much faster than the earlier approaches.