【摘要】 随着高比例可再生能源的接入，配电网电压越限、网络损耗增大和风光消纳等问题日益突出。为此，提出一种考虑风光荷联合场景的配电网动态重构方法。首先，针对风光出力和负荷需求的时序特征，提出一种信息最大化GAN（生成对抗网络）方法，用于获得源荷联合场景。接着，构建考虑网络损耗、节点电压偏差、弃风弃光成本的配电网多目标动态重构模型。然后，为提高模型求解效率，采用Fisher时段划分方法确定最佳重构时段范围，并通过NSGA-Ⅱ（二代非支配排序遗传算法）和模糊综合评价方法得到最优重构策略。最后，通过改进的IEEE 33节点配电系统进行仿真分析，验证了所提方法的有效性。更多还原

【Abstract】 The increasing integration of high-penetration renewable energy has led to prominent challenges in distribution networks, including voltage violations, increased network losses, and wind and photovoltaic(PV) power accommodation. To address these issues, this paper proposes a dynamic reconfiguration method for distribution networks that accounts for wind-PV-load scenarios. First, based on the temporal characteristics of wind/PV outputs and load demand, an information-maximizing generative adversarial network(InfoMax-GAN) is introduced to generate source-load scenarios. Subsequently, a multi-objective dynamic reconfiguration model is formulated, incorporating network losses, node voltage deviations, and wind/PV curtailment costs. To enhance computational efficiency, Fisher optimal partition is employed to determine optimal reconfiguration intervals. The non-dominated sorting genetic algorithm II(NSGA-II) and fuzzy comprehensive evaluation method are utilized to derive the optimal reconfiguration strategy. Finally, simulation analysis of a modified IEEE 33-bus distribution system demonstrates the effectiveness of the proposed method.更多还原