配电网动态无功优化方法研究

Research on Dynamic Reactive Power Optimization

针对动态无功优化问题的难点,通过负荷曲线动态分段来解决控制设备动作次数约束问题,基于当前分段方案和各时段的无功优化方案,对各时段负荷极值点进行潮流计算,根据各节点电压情况动态调整负荷曲线分段数,直到满足设备动作次数约束和节点电压约束。采用改进粒子群算法对负荷曲线分段问题和各时段的静态无功优化问题进行求解,定义全局趋优性能和局部趋优性能,并由此设计自适应惯性权重和加速因子计算公式,在迭代前期和后期分别采用阶段性初始化和变异操作,有效提高了算法的收敛速度和最优解质量。通过对某31节点系统仿真,结果表明了该算法的有效性。

Based on the difficulty of dynamic reactive power optimization,the constraint problems of reactive control equipments’ actions were solved through dynamic segmentation of the load curve,according to the current segmentation scheme and reactive power optimization scheme to take power flow compute for the load extreme value points of every segmentation.The section number was adjusted dynamically by each node voltage,until meet the constraints of equipments’ actions and bus voltage.An improved particle swarm optimization was proposed to solve the problems of load curve divided and static reactive power optimization.Based on the global optimal performance and local optimal performance,the formula of adaptive weights and the acceleration factor were designed.Initialization and mutation of stage operations were adopted respectively in early and late stages,convergence rate and quality of the results were enhanced effectively.A 31-bus radial distribution system showed the effectiveness of the proposed methods in the paper.