基于不同算法求解子问题的Benders分解法在无功规划中的应用  

Application of Benders Decomposition Based on Different Method to Solve Sub-problems in Reactive Power Planning

作　　者：黄婷[1] 黄伟[1] 

机构地区：[1]华北电力大学电气与电子工程学院,北京102206

出　　处：《陕西电力》2013年第3期23-26,共4页Shanxi Electric Power

摘　　要：利用Benders分解法将复杂的非线性无功规划问题分为投资子问题和运行子问题,通过约束割集将子问题联系在一起,在2个子问题间迭代求解,经过有限次迭代,最终求出最优解。求解子问题时,运用DantzigWolfe分解法求解运行子问题,运用改进单纯形法求解投资子问题,并考虑了多种运行方式下最优的无功优化方案。This paper decomposes the complex problem of reactive power planning into investment sub-problem and operational subproblem using the Benders decomposition method. The sub-problems are associated with the Benders cut. And ultimately find the optimal solution after a finite number of iterations between the two sub-problems. Dantzig-Wolfe decomposition method is used to solve the operation sub-problem, and revised simplex method is used to solve the investment sub-problem. And the optimal reactive power planning program is considered under different operating conditions.

关 键 词：无功规划 BENDERS分解法 Dantzig—Wolfe分解法 改进单纯形法 多运行方式 

分 类 号：TM91[电气工程—电力电子与电力传动]