Statistical Bus Ranking for Flexible Robust Unit Commitment
Statistical Bus Ranking for Flexible Robust Unit Commitment
复制标题
DOI：
10.1109/tpwrs.2018.2864131
复制
发表时间：
2019-01-01
影响因子：
6.6
通讯作者：
Anderson, C. Lindsay
中科院分区：
文献类型：
--
作者：
Gupta, Amandeep;Anderson, C. Lindsay
随着全球电力系统中不确定可再生能源容量的增加，工业和学术研究人员都在寻求一种可扩展的，透明的，有效的方法来解决不确定性下的机组承诺问题。本文提出了一种统计排序方法，允许自适应鲁棒随机机组组合使用模块化结构，急需的灵活性。具体而言，本文介绍了一种总线排名方法，确定最关键的总线的基础上最坏的情况下的度量。一个重要的创新是能够识别替代指标，排名的不确定性集，例如，以最小化经济调度成本或斜坡的需要，提供一个定制的强大的单位承诺的解决方案。与传统的鲁棒机组组合模型相比，该模型将统计工具与电力系统网络分析框架相结合。所得到的配方是容易实施和定制的系统操作员的需要。该方法及其应用对其他已建立的方法进行了验证，表现出等效的解决方案的国家的最先进的方法。在IEEE-30、IEEE-118和pegase-1354网络上进行了案例研究。此外，巴士排名制定的灵活性说明通过实施的最坏情况下的度量的替代定义。结果表明，总线排名方法执行以及最好的这些方法，提供额外的灵活性和潜在的并行化。
As the level of uncertain renewable capacity increases on power systems worldwide, industrial and academic researchers alike are seeking a scalable, transparent, and effective approach to unit commitment under uncertainty. This paper presents a statistical ranking methodology that allows adaptive robust stochastic unit commitment using a modular structure, with much-needed flexibility. Specifically, this paper describes a bus ranking methodology that identifies the most critical buses based on a worst-case metric. An important innovation is the ability to identify alternative metrics on which to rank the uncertainty set-for example, to minimize economic dispatch cost or ramping needs, to provide a customized robust unit commitment solution. Compared to traditional robust unit commitment models, the proposed model combines statistical tools with analytical framework of power system networks. The resulting formulation is easily implementable and customizable to the needs of the system operator. The method and its applications are validated against other established approaches, showing equivalent solution to the state-of-the-art approach. Case studies were conducted on the IEEE-30, IEEE-118, and the pegase-1354 networks. In addition, the flexibility of bus ranking formulation is illustrated through implementation of alternative definitions of worst-case metrics. Results show that the bus ranking method performs as well as the best of these methods, with the provision of additional flexibility and potential for parallelization.