数据驱动的配电网电压控制方法研究

Research on data-driven voltage control method for distribution networks

  • 摘要: 在分布式电源大量接入的背景下,配电网的电压运行状态呈现出更加频繁、且难以人工判断的短周期扰动特征,传统依赖本地量测和固定动作逻辑的调节方式已难以在复杂场景下保持足够的响应速度与稳定性。围绕“预测优化控制”双层结构展开研究:首先基于运行数据构建短时电压预测模型,利用历史序列和局部空间信息预测下一时段电压趋势;随后在预测基础上,设计带约束的优化控制模型,将电压偏差、动作代价和设备能力纳入统一决策框架;最后通过算例验证该模型在典型馈线场景中的预测能力、控制效果及稳定性表现。

     

    Abstract: Against the backdrop of the rapid influx of distributed power sources, the voltage operation status of distribution networks has shown more frequent and difficult-to-manually judge short-period disturbance characteristics. The traditional regulation method relying on local measurement and fixed action logic has become difficult to maintain sufficient response speed and stability in complex scenarios. This paper focuses on the "predictive - optimized control" two - layer structure for research: a short - term voltage prediction model is constructed based on operational data, and the voltage trend in the next period is characterized by historical sequences and local spatial information. Subsequently, a constrained optimization control model is designed based on the predictors, and voltage deviation, action cost and equipment capacity are incorporated into the unified decision - making framework. Finally, the predictive ability, control effect and stability performance of the model in typical feeder scenarios are verified through numerical

     

/

返回文章
返回