基于数据驱动知识显式嵌入的配电网最优需求响应策略

Optimal Demand Response Strategy for Distribution Network Based on Data-driven Knowledge Explicit Embedding

充分挖掘多元需求侧资源的灵活性,对于提升分布式新能源广泛接入背景下的新型配电网运行可靠性和经济性具有重要意义。然而,目前关于需求响应策略的解析化方法大多基于较为理想的用户行为观测和参数假设,纯数据驱动方法难以兼顾电网侧运行的复杂约束,策略的可用性存疑。为此,文章提出基于数据驱动知识显式嵌入的需求响应策略,首先,考虑到需求侧资源灵活性的强时段耦合特性,提出需求侧资源动态模型,定量分析需求侧灵活性资源的响应特性;其次,提出数据驱动知识的显式解析方法,将需求侧灵活性描述为混合整数线性模型并嵌入至配电网优化运行模型中,实现灵活实用的新型配电网供需交互与协调运行。最后,通过仿真算例验证所提方法兼具解析模型和数据驱动方法的优势,为不完全信息观测条件下源网荷协调运行提供较为实用化的解决方案。

The full exploitation of the flexibility of diverse demand-side resources is of great significance for enhancing the reliability and economic efficiency of the operation of the new distribution network under the background of widespread access to distributed new energy.However,current analytical methods for demand response strategies are mostly based on ideal user behavior observations and parameter assumptions,while pure data-driven methods are difficult to balance the complex constraints of grid-side operations,raising questions about the availability of strategies.To address this issue,this paper proposes a demand response strategy based on data-driven knowledge explicitly embedded.Firstly,considering the strong temporal coupling characteristics of the flexibility of demand-side resources,a dynamic model of demand-side resources is proposed to quantitatively analyze the response characteristics of flexible resources on the demand side.Secondly,an explicit analytical method for data-driven knowledge is proposed to describe the flexibility on the demand side as a mixed-integer linear model and embed it into the optimization operation model of the distribution network,achieving flexible and practical supply-demand interaction and coordinated scheduling of the new distribution network.Finally,through simulation examples,the advantages of the proposed method,which combines analytical models and data-driven methods,are verified,providing a more practical solution for the coordinated operation of generation,transmission,and load under incomplete information observation conditions.