面向规模化分布式可再生能源并网的群网协同互动优化方法

Cluster-Grid Cooperative Interaction Optimization Method for Large-scale Distributed Renewable Energy Integration

实现规模化分布式可再生能源接入下电网的稳定运行,是低碳化转型新形势下电网优化运行的重要研究方向。针对分布式可再生能源聚合形成集群参与电网优化运行带来的多主体性特点以及随机性问题,提出面向规模化分布式可再生能源并网的群网协同互动优化方法。首先,通过在灵活性资源的约束空间中构建高维内接长方体,形成表征灵活调节范围的可调功率域,通过构造可调功率域对可再生能源集群波动范围全覆盖的灵活性约束,使优化结果在可再生能源集群随机波动范围内全场景可行。然后,基于电网运行多主体的自利自主特性,构建满足同质个体并行优化、非同质个体串行优化要求的势博弈群网协同互动优化模型,提出与模型相适应的分布式优化方法与流程。最后,仿真算例表明,所提方法有效实现了多主体协同互动优化,并保障了优化结果的全场景可行。

The stable operation of the power grid with large-scale distributed renewable energy integration is an important research direction for optimizing the operation of the power grid under the new situation of low-carbon transformation.In response to the multi-agent characteristics and randomness issues brought about by the participation of the clusters formed by distributed renewable energy aggregation in optimizing power grid operation,an optimization method for cluster-grid cooperative interaction for largescale distributed renewable energy integration is proposed.First,by constructing a high-dimensional inscribed rectangle in the constraint space of flexible resources,an adjustable power range is formed to represent the flexible adjustment range.By constructing a flexible constraint of adjustable power domain to fully cover the fluctuation range of renewable energy clusters,the optimization results are feasible in all scenarios within the random fluctuation range of renewable energy clusters.Then,based on the self-interested and autonomous characteristics of multi-agents in power grid operation,an optimization model of potential game cluster-grid cooperative interaction is constructed to meet the requirements of parallel optimization for homogeneous individuals and serial optimization for non-homogeneous individuals.A distributed optimization method and its process suitable for the model are proposed.Finally,simulation cases show that the proposed method effectively achieves multi-agent collaborative interaction optimization and ensures the feasibility of the optimization results in all scenarios.