预条件处理CG法大规模电力系统潮流计算

The Preconditioned CG Method for Large Scale Power Flow Solution

研究了预条件处理的CG(ConjugateGradient)法求解大规模电力系统潮流方程的问题。采用预处理CG法代替传统的LU直接法对高维稀疏潮流方程进行求解,详细比较各种预条件处理技术对CG法潮流方程求解的效果,提出一种新的节点优化排序的IncompleteCholesky预处理方法,实验分析证明它是CG法快速求解潮流的一种十分有效的预处理方法。对IEEE-30、IEEE-118和多个合成的大规模电力系统进行潮流计算,结果表明:这种预处理方法比其它预处理方法需要更少的迭代次数和浮点运算次数,对超大规模电力系统潮流问题也比传统LU直接法更具速度和存储优势。在电力系统互联程度不断增加使其潮流计算面临大规模甚至超大规模计算压力时,该方法能够成为传统方法的一个替代。

This paper presents a detailed investigation into the problem of the preconditioned CG method for Large-scale power flow solution. The preconditioned CG method was used instead of the traditional LU direct method for solution of the large sparse sets of linear equations. Studies comparing the performance of kinds of preconditioner have been preformed. An Incomplete Cholesky preconditioner based on a novel ordering scheme has been presented which has been proved more effective for using CG method for power flow computation. Tests have been performed with networks of IEEE Test Systems and synthesized bulk systems, and shown the better effectiveness in computation for the proposed method than other preconditioners, even so for super large-scale power system comparing with LU direct method.