基于文本Embedding和相似度计算的FA范围准确性校核方法研究  

Research on FA Range Accuracy Verification Method Based on Text Embedding and Similarity Calculation

作　　者：赵成睿 李斌[1] 李洪全[2] 张照芳[2] 蒋熙蕴 ZHAO Chengrui;LI Bin;LI Hongquan;ZHANG Zhaofang;JIANG Xiyun(School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing,Jiangsu 210000,China;Chengxian College,Southeast University,Nanjing,Jiangsu 210000,China)

机构地区：[1]南京工程学院电力工程学院,江苏南京210000 [2]东南大学成贤学院,江苏南京210000

出　　处：《自动化应用》2024年第20期15-17,21,共4页Automation Application

摘　　要：在现代电力系统核心运营中,馈线自动化扮演着提高供电可靠性和服务质量的核心角色。针对FA系统故障定位的精确性挑战,提出了一种基于文本向量化和相似度技术的创新校核方法。与传统手动校核流程相比,该方法显著提升了故障诊断的准确度和操作效率。此项研究为电力系统故障响应与维护流程提供了一种高效、可靠的技术解决方案,对确保电网运行的稳定性和安全性具有极为重要的意义。In the core operation of modern power systems,feeder automation(FA)plays a core role in improving power supply reliability and service quality.Aiming at the accuracy challenge of FA system fault location,this study proposes an innovative verification method based on text vectorization and similarity technology.Drawing on the traditional manual calibration process,this method significantly improves the accuracy of fault diagnosis and operational efficiency.This research provides a reliable technical solution for power system fault response and maintenance processes,which is great significance for ensuring the stability,efficiency and safety of power grid operations.

关 键 词：馈线自动化 故障分析 准确性校核 大型语言模型 文本向量化 相似度计算 

分 类 号：TM76[电气工程—电力系统及自动化]