稳定域概念下考虑保守性的电力系统在线暂态稳定评估方法

发布时间：2018-04-02 01:26

本文选题：在线暂态稳定分析　切入点：稳定域　出处：《电网技术》2016年04期

【摘要】：电力系统故障后在线暂态稳定评估是一个难题。目前基于数据挖掘方法的相关研究存在2个问题:评估规则缺乏电力系统理论依据,评估结果难以保证保守性。针对第1个问题,文章提出稳定域概念下的暂态稳定评估方法,其稳定评估规则是对稳定边界的近似与变换。针对第2个问题,文章提出一种基于新型支持向量机的暂态稳定规则训练算法。该算法的核心在于通过调整传统支持向量机模型得到保守支持向量机和激进支持向量机,利用这2种支持向量机生成的边界将输入空间分为3个区域:稳定区域、不稳定区域和灰色地带。稳定区域和不稳定区域的稳定性评估确保正确,灰色地带尽量小。最后,分别在IEEE 39节点系统和某省级电网中验证了该评估方法的有效性。测试结果不存在漏报警和误报警,灰色地带随着输入特征数的增加而减小,根据需要可以控制在一定范围内。
[Abstract]:Online assessment of power system transient stability after fault is a problem. At present, there are 2 problems related to research based on data mining methods: evaluation rules based on the lack of power system theory, the evaluation results to ensure the conservation. For the first problem, the transient stability domain under the concept of assessment methods, the stability assessment is similar to the rules the stability boundary and transform. For the second problem, this paper proposes a new training algorithm for transient stability rules based on support vector machine. The core of the algorithm is conservative and radical support vector machine support vector machine by adjusting the traditional support vector machine model, using the 2 kinds of support vector machine to generate the input space boundary points for the 3 regions: stable region and the unstable region gray area. The stable region and unstable region stability assessment to ensure that the correct, gray area as much as possible Finally, the effectiveness of the evaluation method is verified in the IEEE 39 node system and a provincial power grid. The test results do not leak alarm and false alarm. The gray area decreases with the increase of the number of input features, and it can be controlled within a certain range according to the need.

【作者单位】：电力系统及发电设备控制和仿真国家重点实验室(清华大学电机系);
【基金】：国网湖南省电力公司科技项目(SGHN0000DKJS 1300221)
【分类号】：TM712

【相似文献】