CHEN YONGJIN. A Coordinated Optimization for Active Distribution Network Reconfiguration and Volt/var Optimization Based on Mixed Integer Convex Programming. [J]. Power capacitor & reactive power compensation, 2020, 41(6): 21-29.
CHEN YONGJIN. A Coordinated Optimization for Active Distribution Network Reconfiguration and Volt/var Optimization Based on Mixed Integer Convex Programming. [J]. Power capacitor & reactive power compensation, 2020, 41(6): 21-29. DOI: 10.14044/j.1674-1757.pcrpc.2020.06.004.
A Coordinated Optimization for Active Distribution Network Reconfiguration and Volt/var Optimization Based on Mixed Integer Convex Programming
摘要
传统方法将有源配电网重构和无功电压优化视为相对独立的子问题研究
缺乏将两者协调考虑。本文提出一种有源配电网重构与无功电压协调优化的混合整数凸规划模型
并采用凸优化方法获取最优解
从而解决配电网的网架重构和无功电压优化的协调运行问题。首先
在配电网线路潮流方程的基础上
建立有源配电网重构和无功电压协调优化模型。然后
对配电网潮流约束、配电网设备运行约束和辐射状配电网网架约束进行凸变换处理
将原非凸模型转化为具有凸特性的混合整数二阶锥规划(mixed integer second order conic programming
MISOCP)模型或者混合整数线性规划(mixed integer linear programming
Active distribution network reconfiguration(DNR) and volt-var optimization(VVO) are usually considered as relatively two independent issues for study
which is lack of coordination consideration of them.In this paper
a mixed integer convex programming-based method for the coordination of DNR and VVO is proposed.The optimal solution is obtained by the use of convex optimal method so to solve the problem of coordination operation of grid reconfiguration of distribution network and reactive voltage optimization.Firstly
a coordinated model for DSR and VVO is set up on the basis of branch flow equations.Then
the constraints of power flow
equipment operation and network frame of distribution network are subject to convex transformation treatment
the original non-convex model is converted into a mixed integer convex programming model with convex feature and the the obtained can be solved by the existing convex optimization software.Finally
simulations are tested on the 33-bus distribution system.The comparison results of 5 operation scenarios have verified that the proposed mixed integer second order conic programming(MISOCP) and mixed integer linear programming(MILP) models are effectively applied to solve the coordinated problem of DNR and VVO and their solutions are the same within the engineering permissible error range.