基于改进K-means聚类计及分布式光伏和电动汽车的园区负荷聚合体的最优构建

Optimal construction of industrial park load polymers concerning the distributed photovoltaic and electric vehicles based on improved K-means clustering

在对用户进行划分的基础上,考虑电动汽车与光伏发电对用户负荷的优化配置,对于实现需求侧资源的高效利用,具有至关重要的意义。因此,提出了一种将K-means聚类算法与万有引力模型相结合的改进聚类模型。首先,采用所提模型对负荷进行聚类;其次,对负荷的特性进行相应的划分,并依据不同的负荷特性,采用一定量的新能源加以匹配从而形成一个负荷聚合体,为了实现负荷聚合体的波动最小、用户效益的最大化,采用相应的优化策略,使得用户效益实现最大化的同时考虑整合需求侧资源;然后,建立兼顾总聚合模型和参与聚合用户效益的多目标优化模型,并运用遗传算法对优化模型进行求解;最后,通过具体算例证实了所提优化与聚合模型的正确性及有效性。

On the basis of user segmentation,the optimal allocation for the user load considering of the electrical vehicles and photovoltaic power generation is of great significance to achieve efficient utilization of resources on the demand side.Therefore,this paper presents a new approach which combines the K-means clustering algorithm with the law of universal gravitation to polymerizing the demand side energy resources of a particular region.Firstly,the proposed model is applied to polymerize the load.Then,after the load polymerization,the corresponding optimal configuration is implemented on the output of photovoltaic as well as power vehicles to form a load polymer,that is,the corresponding control strategy puts great emphasis on maximizing the benefit of the users and integrating the demand side resources.Furthermore,a master-slave optimization model,taking the polymerization model and user benefits after participating in polymerization into consideration,is established and then solved by the improved genetic algorithm.Finally,a practical case is employed to verify the feasibility and effectiveness of the polymerization optimization model.