基于奔德斯算法的安全约束机组组合方法  

A Method for Security Constrained Unit Commitment Based on Benders Algorithm

作　　者：王楠[1] 张粒子[2] 袁喆[3] 张黎明[1] 李雪[1] 

机构地区：[1]天津市电力公司电力科学研究院,天津市西青区300384 [2]华北电力大学电气与电子工程学院,北京市昌平区102206 [3]天津大学电气自动化学院,天津市南开区300072

出　　处：《电网技术》2012年第10期203-208,共6页Power System Technology

摘　　要：针对采用混合整数规划算法直接求解安全约束机组组合,将使计算效率大幅度降低,而利用奔德斯算法求解则存在着算法振荡和受系统规模制约求解效率下降的问题,提出了一种基于奔德斯算法的安全约束机组组合新方法。该方法在奔德斯算法的基础上,通过纳入新增越限约束校正环节,控制了奔德斯割的寻优方向,通过增加起作用约束识别环节,缩小了奔德斯算法的寻优空间,进而提高了安全约束机组组合优化问题的求解效率。6机3节点和54机118节点算例验证了所提方法的有效性。When security-constrained unit commitment（SCUC） was directly solved by mixed integer programming algorithm,the calculation efficiency would considerably decrease,and when SCUC was solved by Benders algorithm,it led to the problem that the solution efficiency would decrease due to the algorithm shock and the restriction of system scale.A new Benders algorithm-based method to solve SCUC was proposed.Based on Benders algorithm,by means of adding the link to correct the constraint of out-of-limit after the iteration of Benders master problem,the search direction of Benders cut could be controlled;by means of adding the link to identify taken-effect constraints the search space of Benders algorithm was reduced,thus the solution efficiency of SCUC optimization could be improved.The effectiveness of the proposed method was verified by simulation results of 6-machine 3-bus system and 54-machine 118-bus system.

关 键 词：电力系统 安全约束机组组合 奔德斯分解 混合整数线性规划 

分 类 号：TM733[电气工程—电力系统及自动化]