基于改进小生境遗传算法的配电网全网无功优化

General Reactive Power Optimization of Whole Distribution Network Based on Improved Niche Genetic Algorithm

提出一种配电网全网无功优化方法,该方法将整个配电网划分为高、中、低压3个等级,各级电网再根据自身特点分解为优化单元,遵循从下至上优化的原则,上级主网优化时计及下级子网无功优化的影响,计算结果更准确可靠。各优化单元的无功优化通过改进的小生境遗传算法来实现,该算法将个体距离判别标准设置为动态函数,能够更灵活、更准确地跟踪个体距离,收敛速度快且能在全局范围内找到最优解。针对某配电网的计算实例验证了配电网全网无功优化方法的优越性。

A reactive power optimization method for whole distribution network is proposed, which divides whole distribution network into three levels, i.e., high voltage level, medium voltage level and low voltage level, in accordance with its own features each sub-network is re-divided into optimum units. According to the principle of optimizing from bottom to top, during the reactive power optimization of network in higher voltage level the influence of reactive power optimization of the sub-network in lower voltage level should be taken into account to make the calculation results more accurate and reliable. The reactive power optimization of each optimum unit is implemented by improved niche genetic algorithm in which the distinguishing criterion of individual distance is set as dynamic function, thus the individual distance can be traced more flexibly and accurately, the convergence is fast and global optimal solution can be found. Calculation results of a certain distribution network validate the superiority of the proposed reactive power optimization method for whole distribution network.