| 引用本文: | 王彬,郭文鑫,向德军,王鹏,余志文.基于改进支持向量机和两步式聚类分析的电网关键断面辨识和精细规则生成方法[J].电力自动化设备,2017,37(9): |
| WANG Bin,GUO Wenxin,XIANG Dejun,WANG Peng,YU Zhiwen.Grid critical section detection and fine operational rule generation based on improved support vector machine and two-step clustering analysis[J].Electric Power Automation Equipment,2017,37(9): |
|
| 摘要: |
| 在一种基于电网运行仿真大数据的知识管理方法框架的基础上，进行改进和实用化，建立电网安全运行知识库。改进了一种用于判断关键断面是否生效的支持向量机分类模型，能够减弱不平衡数据的影响，并降低关键断面漏警概率。提出了一种用于生成不同形式精细规则的层次聚类-Kmeans聚类两步式聚类分析方法，提高了训练的效率。将上述改进知识管理方法在广东电网实际系统中进行算例分析，证明其可行性与有效性。 |
| 关键词: 电网安全运行知识 知识管理 知识库 支持向量机 聚类分析 关键断面 精细规则 |
| DOI：10.16081/j.issn.1006-6047.2017.09.023 |
| 分类号:TM761 |
| 基金项目:中国南方电网公司科技项目(GDKJ00000058) |
|
| Grid critical section detection and fine operational rule generation based on improved support vector machine and two-step clustering analysis |
|
WANG Bin1, GUO Wenxin1, XIANG Dejun1, WANG Peng2, YU Zhiwen1
|
|
1.Guangdong Power Grid Power Dispatching Control Center, Guangzhou 510600, China;2.Beijing Qingda Gaoke System Control Company, Beijing 102208, China
|
| Abstract: |
| For the practical application, the knowledge management framework based on the big operational and simulative data of power system is improved and a knowledge base of power-system safe operation is established. A classification model based on the support vector machine for detecting the critical section of power grid is improved to weaken the impact of unbalanced data and reduce the probability of critical section missing. A two-step analysis based on the hierarchical clustering and Kmeans clustering is proposed for generating the fine operational rules of different types to increase the training efficiency. The improved knowledge management method is applied to a practical power system of Guangdong Power Grid and result proves its feasibility and effectiveness. |
| Key words: power-system safe operation knowledge knowledge management knowledge base support vector machines clustering analysis critical section fine operational rules |
