含分布式电源的配电网潮流改进算法

Improved power flow calculation of distribution network with DG

为了解决传统的配电网潮流算法不能求解含有分布式电源的问题,对传统的前推回代法进行了改进。由于传统的算法不能处理PV型节点,引入节点电抗矩阵后,把PV节点转换为PQ节点进行求解。同时为了解决辐射型配电网传统的无功初值确定方法增加迭代次数的问题,基于无功分摊原理改进了无功初值的确定方法。最后通过对美国PG&E69节点系统建立了仿真模型,得到了配电网单独地并入相应的视作PQ、PI、PV节点的DG以及各类分布式电源混合安装后潮流计算的迭代次数和计算时间的对比结果和基于分摊原理的方法、初值为0以及初值为平均值三种方法的比较结果。通过对结果的分析可以得到:此改进后的算法不论对单个还是混合接入系统的情况都能快速求解;同时新的无功初值确定方法减少了迭代次数,节省了计算时间,验证了改进算法的正确性和可行性。

Because the traditional power flow algorithm can＇t solve the problem of distribution network with DG, the traditional back/forward sweep method is improved. The traditional algorithm can＇t handle PV nodes, so the PV nodes are converted to PQ nodes to solve after introducing node reactance matrix. At the same time, in order to solve the problem that traditional reactive initial value determination method will increase iteration times in radial distribution network, the method of determining the reactive initial value is improved based on the reactive apportion principle. Finally, the simulation model of American PG＆E 69 nodes system is established. Based on this, the iterative times and computation time of power flow calculation are contrasted between two cases： one is to connect the distribution network to corresponding DGs which are regarded as PQ, PI, PV nodes individually, and the other is to install all types of DGs in a mixed way. Then three kinds of methods are compared, namely, the method based on the apportion principle, the method with the initial value as zero and the method with the initial value as average value. Analysis results show that the improved algorithm can calculate the power flow quickly in both cases; meanwhile, the new reactive initial value determination method reduces the iteration times and saves calculation time, which verifies the validity and feasibility of the improved algorithm.