基于内点割平面算法的电力系统机组优化组合研究  

Study on Unit Commitment of Power System Based on Improved Interior Point Cutting Plane Method

作　　者：丁晓莺[1] 王建学[1] 高卫恒[1] 陈晨[1] 

机构地区：[1]西安交通大学电气工程学院,陕西西安710049

出　　处：《陕西电力》2008年第1期20-24,共5页Shanxi Electric Power

摘　　要：合理的开停机方案将为电力系统运行带来很大的经济效益,现代电力系统对机组优化组合算法的收敛速度和解的质量要求越来越高。文章提出了一种改进的内点割平面算法(IPCPM)求解机组组合问题,与原IPCPM相比,在退化解的处理过程中避免了多次的矩阵列交换和秩计算,同时省略了对偶最优基的判断流程,使算法计算量明显减少,速度更快。通过多个不同规模和特性的系统测算,表明该算法能有效地产生割平面约束,计算结果准确合理,改进后的算法实现更方便,计算更快速,适合求解大型电力系统机组组合问题。A rational generation schedule can bring obvious economic benefit for power system operation. Modem power system requires higher convergence speed and better solution quality of the generation scheduling algorithm. In this paper an improved interior point cutting plane method （IPCPM） is proposed to solve the unit commitment （UC）problem. Compareing with the original IPCPM, a number of column exchanges and rank calculations are avoided in the degenerated solution treatment and the calculation process of dual optimal basis is elided, which causes less computation burden and faster calculation speed of the algorithm. Numerical simulations on several test systems have shown that the presented algorithm can obtain cutting plane constraints effectively and give exact scheduling result. In conclusion, the improved IPCPM is more convenient and fast to execute, and is very suitable for solving large-scale UC problem.

关 键 词：机组组合 内点割平面算法 退化问题 最优基 

分 类 号：TM731[电气工程—电力系统及自动化] TP301.6[自动化与计算机技术—计算机系统结构]