论文标题

偶然受限的OPF:具有机密性保存的分布式方法

Chance-constrained OPF: A Distributed Method with Confidentiality Preservation

论文作者

Jia, Mengshuo, Hug, Gabriela, Su, Yifan, Shen, Chen

论文摘要

鉴于电力系统中风能的百分比增加,因此正在促进偶然受限的最佳功率流(CC-OPF)计算,作为将风力不确定性考虑到保证的安全水平的一种手段。与区域网格中的本地CC-OPF相比,多区域互连网格的全球CC-OPF能够在不同地区进行协调,因此在整合高比例的风能发电比例时提高了经济效率。但是,在这个全球问题中,多个区域独立系统运营商(ISO)参与了决策过程,从而增加了分布式但协调的方法。最值得注意的是,由于法规限制,商业兴趣和数据安全,区域ISO可能拒绝与他人共享机密信息,包括发电成本,负载数据,系统拓扑和线路参数。但是,需要此信息来构建和解决跨越多个区域的全局CC-OPF。为了解决这些问题,本文提出了一种具有机密性保存的分布式CC-OPF方法,该方法使区域ISOS能够在其区域内确定最佳的可分配世代,而无需披露机密数据。此方法不需要参数调谐,也不会面临融合挑战。 IEEE测试用例的结果表明,此方法非常准确。

Given the increased percentage of wind power in power systems, chance-constrained optimal power flow (CC-OPF) calculation, as a means to take wind power uncertainty into account with a guaranteed security level, is being promoted. Compared to the local CC-OPF within a regional grid, the global CC-OPF of a multi-regional interconnected grid is able to coordinate across different regions and therefore improve the economic efficiency when integrating high percentage of wind power generation. In this global problem, however, multiple regional independent system operators (ISOs) participate in the decision-making process, raising the need for distributed but coordinated approaches. Most notably, due to regulation restrictions, commercial interest, and data security, regional ISOs may refuse to share confidential information with others, including generation cost, load data, system topologies, and line parameters. But this information is needed to build and solve the global CC-OPF spanning multiple areas. To tackle these issues, this paper proposes a distributed CC-OPF method with confidentiality preservation, which enables regional ISOs to determine the optimal dispatchable generations within their regions without disclosing confidential data. This method does not require parameter tunings and will not suffer from convergence challenges. Results from IEEE test cases show that this method is highly accurate.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源