| 引用本文: | 叶远波,李端超,谢 民,王志华,王骏东.基于知识图谱的二次设备测试自动配置方法[J].电力系统保护与控制,2022,50(12):162-171.[点击复制] |
| YE Yuanbo,LI Duanchao,XIE Min,WANG Zhihua,WANG Jundong.Automatic configuration method of secondary equipment test based on a knowledge graph[J].Power System Protection and Control,2022,50(12):162-171[点击复制] |
|
| 摘要: |
| 随着智能变电站的发展，二次设备更新换代频繁且智能化程度不断提升，以往采用手动测试配置的方法已难以满足工作需求。提出一种基于知识图谱的二次设备自动配置方法。首先，模拟常规测试核心流程，构建基于测试任务的信息点、基于设备厂家的信息点检索路径以及基于测试配置信息类型的命名实体知识三类知识图谱。然后，根据测试设备及测试功能，利用所构建的知识图谱检索需配置信息点，确定设备厂家并基于子图匹配信息路径。采用改进的BERT模型计算配置信息点与描述文本之间的相似性，从而完成映射配置。此外，还加入了端口地址配置数据字符串校验。实验仿真结果表明，基于知识图谱的二次设备测试自动配置方法合理有效，并具有极高的准确性。 |
| 关键词: 二次设备测试 知识图谱 BERT模型 文本向量 路径推理 |
| DOI：DOI: 10.19783/j.cnki.pspc.211193 |
| 投稿时间：2021-08-30修订日期：2021-12-21 |
| 基金项目:国家电网公司科技项目资助(5212002000AR) |
|
| Automatic configuration method of secondary equipment test based on a knowledge graph |
| YE Yuanbo,LI Duanchao,XIE Min,WANG Zhihua,WANG Jundong |
| (1. State Grid Anhui Electric Power Co., Ltd., Hefei 230022, China; 2. Wuhan Kemov Electric Co., Ltd., Wuhan 430223, China) |
| Abstract: |
| With the development of smart substations, the replacement of secondary equipment is frequent and intelligence is increasingly improved. The former manual test configuration method finds it difficult to meet the work needs. An automatic configuration method for secondary equipment based on a knowledge graph is proposed. First, the core process of conventional testing is simulated to build a knowledge graph based on three types of data, test task information points, information point retrieval paths of equipment manufacturers, and named entity knowledge of test configuration information types. Then, from the test equipment and testing capabilities, the information points needed to be configured are retrieved by the constructed knowledge graph, the device manufacturer is determined and the information path is matched based on the subgraph. The improved BERT model is used to calculate the similarity between the configuration information points and the description text to complete the mapping configuration. In addition, a string verification of port address configuration data is added. The simulation results show that the method based on the knowledge graph is reasonable and effective and has high accuracy.
This work is supported by the Science and Technology of State Grid Corporation of China (No. 5212002000AR). |
| Key words: secondary equipment test knowledge graph BERT model text vector path reasoning |
