基于局部几何参数化和JFNG算法的微电网群分布式连续潮流计算

Distributed continuous power flow calculation of microgrid group based on local geometric parameterization and JFNG algorithm

由于分布式电源、电动汽车等新型电气化负荷内存在无功限幅约束,传统连续潮流在分析微电网的带负荷能力和微电网之间的联络能力时存在因错误识别节点类型而导致计算失败的现象,并且其不适配于微电网群的分布式管理模式。提出一种局部几何参数化连续潮流算法,相较于弧长参数化,该算法具有参数化方程易于解耦、分岔点计算精度较高等优点;采用费舍尔伯明斯特互补函数来处理非光滑无功限幅特性,避免节点类型频繁切换的问题。基于连续潮流模型,结合非精确牛顿-广义最小残差(JFNG)算法分布式框架,提出仅交换边界信息就可收敛的微电网群分布式连续潮流算法。算例结果验证了所提算法的有效性。

Since the reactive power amplitude limiting constraints is existed in the new electrification loads such as distributed generation and electric vehicles,the phenomenon of calculation failure caused by wrong identification of node types is existed when the load capacity of microgrid and connection ability between microgrids are analyzed by the traditional continuous power flow,which is not suitable for the distributed management mode of microgrid group.A local geometric parametric continuous power flow algorithm is pro⁃posed,it has the advantages such as easy decoupling of parametric equation and high calculation accuracy of bifurcation point compared with the arc length parameterization.Fischer-Burmeister complementary func⁃tion is adopted to deal with the non-smooth reactive power amplitude limiting characteristics,and the frequent switching of node types is avoided.Based on the continuous power flow model and distributed framework of Jacobian-free Newton-generalized minimum residual(JFNG)algorithm,a distributed continuous power flow algorithm of microgrid group is proposed,which can converge only by exchanging boundary information.The effectiveness of the proposed algorithm is verified by the example results.