一种牛顿潮流算法收敛性定理的应用研究

牛顿拉夫逊(NR)潮流算法是电力系统潮流计算最常用的算法之一,由于NR法具有平方收敛的特性,收敛速度较快,但它对初值选取的要求较高,初值选取对NR法的收敛性能有很大的影响。针对这一问题,分析了一种牛顿潮流算法的收敛性定理,运用此定理可解决因为初值选取不当而造成的潮流发散问题。此外,该定理可进一步评估选择的初值能否保证NR法收敛,当潮流收敛时,就可以继续进行潮流计算。反之,可以对选择的初值进行适当的调整,如此可以防止因为初值选取不当而造成冗余潮流计算乃至病态潮流问题。通过对IEEE14、IEEE30节点系统和东北电网的仿真分析,验证了定理的可行性和合理性,提高了NR法在潮流计算中的实用性。

A Study on the Application of Convergence Theorem of Newton Power Flow Algorithm

Newton Raphson(NR)power flow algorithm is one of the most commonly used algorithms for power flow calculations,NR method has the property of square convergence,so it converges faster,but it has higher requirements on the selection of initial value.The selection of initial value has a great impact on the convergence performance of the NR method.To solve this problem,a convergence theorem of Newton power flow algorithm is analyzed,which can be applied to solve the problem of power flow divergence caused by improper initial value selection.In addition,the theorem goes further to evaluate whether the chosen initial value can guarantee the convergence of NR method,and when the power flow converges,the power flow calculation can continue,and vice versa,the chosen initial value can be adjusted appropriately.In this way,the redundant power flow calculation and ill conditioned power flow problem caused by improper initial value selection can be prevented.Through further simulation and analysis of IEEE14,IEEE30 nodes system and Dongbei power grid,the viability and rationality of the theorem are verified,and the practicability of NR method in power flow calculation is improved.