基于分解的多目标进化算法的含MMC-HVDC交直流混合系统最优潮流研究

Study on Optimal Power Flow for AC/DC Hybrid System Incorporating MMC-HVDC Based on MOEA/D

提出一种基于分解的多目标进化算法(MOEA/D)的含模块化多电平换流器型高压直流输电(MMC-HVDC)交直流混合系统多目标最优潮流算法,协调统筹系统的经济性和电压质量。首先,基于考虑换流器损耗的模块化多电平换流器(MMC)换流站一般模型,建立目标函数为网络损耗和电压偏差的含直流潮流控制器(DCPFC)交直流混合系统最优潮流数学模型,推导含DCPFC的直流电网MMC换流站采用下垂控制时的雅可比矩阵计算公式。其次,利用MOEA/D对交直流模型加以优化求解得到帕累托(Pareto)最优解集并与改进非劣排序多目标遗传算法(NSGA-2)进行对比。最后,采用熵权法(EWM)对优化结果进行决策支持,选择最佳折中解。以IEEE多节点系统为例,对比两端及三端MMC换流站在不同控制方式下的计算结果,验证了该方法的有效性。

To coordinate system economy and voltage quality,a new multi-objective optimal power flow algorithm based on multi-objective evolutionary algorithm based on decomposition(MOEA/D)was proposed for AC/DC hybrid system incorporating MMC-HVDC.Firstly,based on the general model of MMC converter stations that consider the losses of the converter,the optimal power flow mathematical model of DC power flow controller(DCPFC)containing AC/DC hybrid system with the target functions for network losses and voltage deviations was established.The calculation formula of the Jacobian matrix at the droop control of the DC grid MMC converter stations with DCPFC was derived.Next,the MOEA/D was used to optimize the AC/DC model to obtain the optimal solution set of Pareto and compared with non-dominated sorting genetic algorithm-2(NSGA-2).Finally,according to the entropy weight method(EWM),the optimization results were supported in decision and the best compromise solution was selected.The IEEE multi-node systems were used to compare the calculation results of the two-terminal and three-terminal MMC converter stations under different control modes,which verifies the effectiveness of the method.