RESEARCH INITIATION AWARD: Distributed Implementation of Optimal Power Flow

本研究旨在开发大型电力系统中分布式实现“最优潮流”的新算法。考虑到系统的规模，提出了系统的分解方法。“影子成本”包括区域间有功和无功功率的交换将以这样一种方式指导分布式计算，即经济解释可以放置在区域交换中。区域相对较大，以容纳粗粒度的处理器体系结构。通过将处理过程划分为多个区域，数据通信可以在很大程度上局限于各个区域。区域间通信需求可以大大少于集中实施OPF。然而，分布式处理的主要优点是，由于（分布式）计算能力的增加，解决时间可能会减少。

The proposed research is for developing new algorithms for distributed implementation of "optimal power flow" (OPF) in a large power system. In view of the size of the system a decomposition of the system is proposed. "Shadow costs" on , among other things, interchanges of active and reactive power between regions would guide the distributed calculations in such a way that the economic interpretations can be placed in the regional interchanges. The regions would be relatively large to accommodate a coarse-grained processor architecture. BY dividing the processing into regions, data communication can be largely localized to the regions. The interregional communications requirements can be considerably less than for a centralized implementation of OPF. However, the main advantage of distributed processing is the potential decrease in solution time due to increased (distributed) computing power.