基于风电预测误差聚类的分布鲁棒含储能机组组合

为应对风电不确定性给电力系统调度带来的难题,提出了一种基于风电预测误差聚类的分布鲁棒含储能机组组合模型。首先,基于狄利克雷过程高斯混合模型对风电预测误差进行聚类,建立了数据驱动的风电预测误差模糊集,并进一步建立了考虑风电场间风电预测误差相关性的不确定集。接着提出了考虑储能的分布鲁棒机组组合模型,建立了考虑储能系统循环老化成本的目标函数。针对该模型min-max-max-min的4层结构,将其分解为两阶段问题,在第1阶段中引入运行域变量、爬坡事件约束与储能能量约束,以消去第2阶段中的动态约束,并将第2阶段问题通过KKT条件转化为单层问题,然后采用列约束生成算法对两阶段问题进行求解。最后,通过IEEE 6节点以及IEEE 118节点的算例分析,证明了所提模型的鲁棒性和有效性。

Distributed Robust Unit Commitment with Energy Storage Based on Forecasting Error Clustering of Wind Power

In order to solve the problem of power system dispatching caused by wind power uncertainty, this paper proposes a distributed robust model for unit commitment with energy storage based on forecasting error clustering of wind power. Firstly, based on the Dirichlet process Gaussian mixture model(DPGMM), the wind power forecasting error is clustered to establish a data-driven fuzzy set of wind power forecasting error. The uncertainty set considering the correlation of wind power forecasting errors between wind farms is further established. Then, a distributed robust unit commitment model considering energy storage is proposed, and an objective function considering the cyclic aging of the energy storage system is established. Then the model with the min-max-max-min structure is decomposed into a two-stage problem. In the first stage, the operation region variable, the climbing event constraint and the energy constraint of energy storage are introduced to eliminate the dynamic constraints at the second stage. The second stage problem is transformed into a single-layer problem by KKT condition, and the two-stage problem is solved by the column and constraint generation(C&CG)algorithm. Finally, the robustness and effectiveness of the proposed model are proved by the example analysis based on IEEE 6-bus and IEEE 118-bus systems.