海量异构分布式能源资源接入配电网，将对电力系统“源随荷动”的调度模式造成挑战。聚合分布式资源使其以集群形式参与电力系统调度，实现源网荷互动，既能有效应对大规模分布式资源的接入，也能发挥分布式资源整体效益，进一步提高电网的灵活性和可靠性。鉴于此，本文研究基于可行域投影的分布式资源聚合与协调优化调度方法，构建分布式能源资源集群的功率-成本高维可行域，并以此可行域作为资源集群的聚合模型参与电力系统优化调度。首先，本文提出了基于顶点搜索的高维可行域投影算法。该算法通过构建各凸包表面的顶点集，使得新顶点直接与源表面构成新的凸包表面，进而在历次迭代中更新凸包表面外法向向量，使凸包沿着表面法线方向扩张。另外，该算法采用外法向向量聚类，减小了历次迭代中搜索到的重复顶点，提高了凸包扩张的效率。该算法计算量并不随着凸包维度和迭代次数增加呈指数增长，适用于线性系统中高维可行域的投影计算，可应用于考虑时段耦合约束的分布式资源高维聚合模型的计算。其次，本文针对含分布式资源配电网以聚合形式参与电力系统调度的问题，提出了电网-分布式资源集群协调优化调度模型。基于高维可行域投影算法，将分布式资源集群内部约束投影到与上层电网交互的关口功率、成本变量上，构建反映配电网运行和经济特性的高维可行域。此外，基于场景法考虑了风电、光伏等资源的出力不确定性。建立了电网-分布式资源集群协调优化调度模型，以投影得到的高维凸包作为聚合模型参与电力系统优化调度。最后，本文针对电-气-热综合能源系统，提出了考虑阶梯碳交易机制的综合能源系统聚合与协调优化调度方法。综合能源系统中考虑了电解槽、热电联产等能量转化设备，且接入上层电网、气网，通过阶梯碳交易机制量化系统碳排放成本。综合能源系统协调优化调度模型目标函数为电网、气网、综合能源系统的运行总成本。基于高维可行域投影算法计算电功率-气功率-总成本的高维可行域作为综合系统聚合模型，并参与电网、气网的优化调度。本文工作在分布式资源广泛接入电网的背景下，基于可行域投影理论，将分布式资源集群的运行和经济特性投影刻画为高维凸包，并参与电力系统的优化调度，希望本文工作能为电力系统“源网荷互动”调度模式提供技术支撑。

The access of massive heterogeneous distributed energy resources to the distribution network will pose challenges to the "source-follow-load" dispatching model of the power system. Aggregating distributed resources enables them to participate in power system dispatching in the form of clusters and realize source-grid-load interaction. This can not only effectively cope with the access of large-scale distributed resources, but also bring into full play the overall benefits of distributed resources, further improving the flexibility and reliability of the power grid. In view of this, this paper studies distributed resource aggregation and coordinated optimization scheduling methods based on the projection of feasible region, constructs a power-cost high-dimensional feasible region of distributed energy resource clusters, and uses this feasible region as an aggregation model of resource clusters to participate in power system scheduling.First, this paper proposes a projection algorithm of high-dimensional feasible region based on vertex search. This algorithm constructs the vertex set of each convex hull surface, so that the new vertexes directly form a new convex hull surface with the source surface, and then updates the normal vector outside the convex hull surface in previous iterations, so that the convex hull expands along the surface normal direction. In addition, this algorithm uses outer normal vector clustering to reduce the number of duplicate vertexes searched in previous iterations and improve the efficiency of convex hull expansion. The calculation amount of this algorithm does not increase exponentially with the increase of the convex hull dimension and the number of iterations. It is suitable for the projection of high-dimensional feasible regions in linear systems and can be applied to the calculation of high-dimensional aggregation models of distributed resources considering time period coupling constraints.Second, this paper proposes a power grid-distributed resource cluster coordinated scheduling model to address the problem of distribution networks containing distributed resources participating in power system dispatch in an aggregated form. Based on the projection algorithm of high-dimensional feasible region based on vertex search, the internal constraints of the distributed resource cluster are projected onto the gateway power and cost variables that interact with the upper-layer power grid to construct a high-dimensional feasible region that reflects the operation and economic characteristics of the distribution network. In addition, the uncertainty in the output of wind power, photovoltaic and other resources is considered based on the scenario method. A power grid-distributed resource cluster coordination and scheduling model is established, and the high-dimensional convex hull obtained by projection is used as an aggregation model to participate in the power system scheduling.Final, this paper proposes an integrated energy system aggregation and coordinated scheduling method considering the ladder carbon trading mechanism for the electricity-gas-heat integrated energy system. The integrated energy system considers energy conversion equipment such as electrolyzers and cogeneration, and is connected to the upper power grid and gas grid to quantify system carbon emission costs through a ladder carbon trading mechanism. The objective function of the integrated energy system coordinated scheduling model is the total operating cost of the power grid, gas network, and integrated energy system. Based on the high-dimensional projection of feasible region algorithm, the high-dimensional feasible region of electric power-gas power-total cost is calculated as a comprehensive system aggregation model, and participates in the optimal dispatch of the power grid and gas network.This paper works in the context of widespread access of distributed resources to the power grid. Based on the projection of feasible region, the operation and economic characteristics of distributed resource clusters are projected as high-dimensional convex hulls, and participates in the optimal dispatch of the power system. It is hoped that the work of this paper can provide technical support for the "source-grid-load interaction" dispatching model of the power system.