全文下载：2013060628
王威1， 韩学山2， 许星明1， 王勇2， 车仁飞2
(1．山东科技大学 机电工程系，山东 泰安 271019;
2．山东大学 电气工程学院，山东 济南 250061)
摘要：建立了电容器优化投切的动态规划模型。基于无功就地平衡规则和配电网辐射状运行的特点，并利用广度优先搜索算法将电容器划分为多个等级，在此基础上，划分出动态规划的阶段。采用逆序解法求解动态规划问题，给出了电容器对应子网的定义，对每个阶段的各个子网采用原对偶内点法求解最优决策量的浮点解，以网损最小为目标对浮点解归整。算法每个阶段都以网损最小对该阶段的电容器投切容量的浮点解进行归整，从而使整个过程的电容器的整数解更接近最优整数解。算例结果验证了算法的快速性和准确性。
关键词：配电网； 电容器优化投切； 动态规划法； 广度优先搜索
中图分类号： TM 53文献标志码： A
Dynamic Programming for Optimal Sizing of Capacitor in Distribution Network
WANG Wei1, HAN Xue-shan2, XU Xing-ming1, WANG Yong2, CHE Ren-fei2
(Department of Mechanical and Electronic Engineering，Shandong Science and Technology University，Taian 271019， China； School of Electric Engineering， Shandong University， Ji’nan 250061， China)
Abstract:This paper presents a new dynamic programming model for optimal sizing of capacitor in distribution network. Based on the rule of local balance for reactive power and radial construct of distribution networks, capacitors are divided into many grades by utilizing breadth-first search. Based on theses results, the stages of dynamic programming are given. Inverse solution for dynamic programming is adopted. The sub network of capacitor is defined. For the sub network of every stage, the prime-dual interior point method is applied to get the optimal float resolution of decision variables and the integrating objective is minimum network loss. By integrating the float resolution with the objective of minimum network loss in every stage, the integer resolution of whole process is more accurate. The numerical results show that the proposed method is effective and rapidity.
Key words:distribution network; capacitor optimal sizing; dynamic programming; breadth-first search
收稿日期：2013-01-03
基金项目： 山东省优秀中青年科学家科研奖励基金项目(2011BSB01212).
作者简介： 王威（1974—），男，博士.
