首页|基于FP-Growth算法的直流输电系统阀基电子设备缺陷关联性分析

基于FP-Growth算法的直流输电系统阀基电子设备缺陷关联性分析

扫码查看
换流阀控制设备作为直流输电系统的核心设备,对其阀基电子设备进行缺陷异常分析是保证直流输电系统稳定可靠运行的基础.提出一种基于FP-Growth算法的直流输电阀基电子设备缺陷关联性分析方法.首先基于阀基电子设备的基本结构与原理,采集阀基电子设备缺陷数据;接着对原始数据进行预处理,量化编码后导入FP-Growth算法,通过构建FP-Tree,计算其支持度和置信度,分析阀基电子设备的缺陷特征和影响因素以及各元件之间的关联关系.该方法能高效智能实现对直流输电系统核心设备缺陷的关联分析及故障溯源,为运维人员检修策略的制定提供了理论依据.最后以实际直流输电系统换流阀阀基电子设备缺陷数据仿真算例对所提方法的有效性进行了验证.
Correlation Analysis of Defects of Valve Base Electronic Equipment in HVDC Transmission System Based on FP-Growth Algorithm
The converter valve control equipment is the core equipment of the DC transmission system,and the abnormal analysis of the valve base electronic equipment is the basis to ensure the stable and reliable operation of the DC transmission system.A defect correla-tion analysis method for HVDC valve-based electronic equipment based on FP-Growth algorithm is proposed.Firstly,based on the basic structure and principle of valve-based electronic equipment,the defect data of valve-based electronic equipment are collected.Then,the original data are preprocessed,quantized and encoded,and then imported into the FP-Growth algorithm.By constructing FP-Tree,calcu-lating its support and confidence,the defect characteristics and influencing factors of valve-based electronic equipment are analyzed,as well as the correlation between components.The proposed method can efficiently realize the correlation analysis and fault traceability of the core equipment defects of the HVDC transmission system,and provide a theoretical basis for the formulation of maintenance strate-gies for operation and maintenance personnel.Finally,the effectiveness of the modified method is verified by a simulation example of the defect data of converter valve base electronic equipment of actual DC transmission system.

DC transmission systemvalve base electronicsFP-Growth algorithmdefects correlation analysis

肖耀辉、余俊松、李为明、薛海平、王永平、戴剑丰

展开 >

中国南方电网有限责任公司超高压输电公司检修试验中心,广东 广州 510000

南京南瑞继保电气有限公司,江苏 南京 211102

南京邮电大学自动化学院,人工智能学院,江苏 南京 210023

直流输电系统 阀基电子设备 FP-Growth算法 缺陷关联性分析

国家自然科学基金项目

51877037

2024

电子器件
东南大学

电子器件

CSTPCD
影响因子:0.569
ISSN:1005-9490
年,卷(期):2024.47(4)