基于特征点集的分布式能源调度灵活性聚合方法  

An aggregation method for flexibility enhancement in distributed energy scheduling based on feature point sets

作　　者：樊鹏阳 翟超 何宇俊 刘沅昆 张和洪 Pengyang FAN;Chao ZHAI;Yujun HE;Yuankun LIU;Hehong ZHANG(School of Automation,China University of Geosciences,Wuhan 430074,China;Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complea Systems,Wuhan 430074,China;Engineering Research Center of Intelligent Technology for Geo-Eaploration,Ministry of Education,Wuhan 430074,China;Department of Electrical Engineering and Applied Electronics,Tsinghua University,Beijing 100084,China;College of Computer and Data Science,Fuzhou University,Fuzhou 350108,China)

机构地区：[1]中国地质大学(武汉)自动化学院,武汉430074 [2]复杂系统先进控制与智能自动化湖北省重点实验室,武汉430074 [3]地球探测智能化技术教育部工程研究中心,武汉430074 [4]清华大学电机工程与应用电子技术系,北京100084 [5]福州大学计算机与大数据学院,福州350108

出　　处：《中国科学:信息科学》2025年第2期372-387,共16页Scientia Sinica(Informationis)

基　　金：中国地质大学(武汉)“地大学者”人才岗位科研启动经费(批准号:2020138)资助项目。

摘　　要：随着微电网和虚拟电厂等分布式能源管理方式的变革,大量的分布式能源给电力系统的运营商带来了技术挑战并且增大了管理复杂度.一个可控分布式能源的聚合运行模型,可以方便高效地管理大量的分布式能源.本质上,聚合运行模型的可行域是所有分布式能源的多面体可行域的Minkowski和.然而,两个任意多面体的Minkowski和的计算是一个不确定的多项式时间难题.为了精确地刻画并提高大量分布式能源的聚合灵活性,首先,本文提出一种通用的凸约束空间中的Minkowski和计算方法.采用离散点集的凸包逼近真实约束空间,再使用Minkowski和对离散点集进行叠加,并提出离散点集优化算法,使其尽可能地逼近真实约束空间.这种点集是高维度、抽象的,但却表征了分布式能源的运行特征,我们将其定义为特征点集.然后,提出了一种特征点集的选取规则,能够提高调度灵活性,增加调峰能力并减少Minkowski和的运算复杂度.此外,提出一种简化特征点集的算法并且保证了不损失约束空间.最后,从理论上阐释了特征点集法在电网调度中的合理性.数值仿真结果表明,该方法在提高系统聚合灵活性和计算效率方面具有很大潜力.With the transformation of distributed energy management methods such as microgrids and virtual power plants,the proliferation of distributed energy resources presents technical challenges.It increases management complexity for power system operators.A controllable aggregated operation model of distributed energy resources can facilitate the efficient management of many distributed energy resources.Essentially,the feasible region of the aggregated operation model is the Minkowski sum of the polyhedral feasible regions of all distributed energy resources.However,calculating the Minkowski sum of two arbitrary polyhedra is an NPhard problem.To accurately characterize and enhance the aggregation fexibility of numerous distributed energy resources,this paper firstly proposes a method for calculating the Minkowski sum in a general convex constraint space.The method uses the convex hull of a discrete point set to approximate the real constraint space and then overlays the discrete point sets using the Minkowski sum.A discrete point set optimization algorithm is proposed to approximate the real constraint space as closely as possible.This point set is high-dimensional and abstract,but it represents the operational characteristics of distributed energy resources,which we define as the characteristic point set.Next,a selection rule for the characteristic point set is proposed,which can improve scheduling flexibility,increase peak-shaving capability,and reduce the computational complexity of the Minkowski sum.Furthermore,an algorithm for simplifying the characteristic point set is proposed,ensuring the constraint space is not compromised.Finally,the theoretical rationale for using the characteristic point set method in grid scheduling is explained.Numerical simulation results demonstrate that this method has great potential in enhancing system aggregation flexibility and computational eficiency.

关 键 词：分布式能源 特征点集 聚合和分解 Minkowski和 内近似 凸包 

分 类 号：TM73[电气工程—电力系统及自动化]