基于相似搜索与多开端重组的配电网两时间尺度无功协调控制

规模化的分布式电源接入配电网后造成了电压高间歇性波动，影响了无功补偿设备的调节效果。提出基于相似搜索与多开端重组的多目标无功优化算法，考虑了非平滑补偿设备的日投切次数限制，加入变化的次数成本约束，调节尺度为1 h；DG和SVG的实时平滑调节设备，调节尺度为5 min，进行超短期预测反馈调节，优化目标为电压偏差、网损和调控成本综合最优。求解算法采用改进的多目标粒子群算法，运用相似搜索技术，调用适用性最高的历史无功方案作为方案粒子加快算法的初期收敛速度；引用Pareto熵协调多目标间的适应度，同时依托Pareto解集存优能力，在算法收敛后期设计了多开端重组方法，进一步提高优化效果。经实际算例验证了该方法的有效性。

Reactive Power Coordination Control Method of Distribution Network in Two-Time Scale Based on Similar Search and Multi-Start Recombination

The large-scale access of distributed power generation to the distribution network causes high random fluctuations of voltage, which affect the regulation effect of reactive power compensation equipment. Therefore, a multi-objective reactive optimization algorithm based on similar search and multi-start recombination is proposed in this paper. In the optimization model, the daily switching limit of the non-smoothing compensation device is considered, and the changed cost is added. The time scale of the adjustment is 1h. DG and SVG are used to realize prediction and feedback adjustment with the adjustment scale of 5 minutes. The objective of optimization is to minimize voltage deviation, network loss and regulation cost. The improved multi-target particle swarm optimization algorithm is adopted with the similar search technology. And the most applicable historical reactive power scheme is also invoked as the scheme particle to accelerate the initial convergence. Additionally, the improved Pareto entropy is introduced to coordinate the fitness between multiple targets of the optimization model. Meanwhile, relying on the Pareto solution set storage capacity, the multi-start recombination method is proposed to improve the optimization effect of the scheme. Finally, an example is given to verify the effectiveness of the algorithm in solving practical problems.