自动化与仪器仪表2024,Issue(6) :127-131.DOI:10.14016/j.cnki.1001-9227.2024.06.127

基于贝叶斯概率更新的电力物资质量检测方法

A Quality Inspection Method for Electric Power Materials Based on Bayesian Probability Update

苏彦龙 胡雨 计昊 王玉玲 潘士通 刘膨源
自动化与仪器仪表2024,Issue(6) :127-131.DOI:10.14016/j.cnki.1001-9227.2024.06.127

基于贝叶斯概率更新的电力物资质量检测方法

A Quality Inspection Method for Electric Power Materials Based on Bayesian Probability Update

苏彦龙 1胡雨 2计昊 3王玉玲 2潘士通 2刘膨源2
扫码查看

作者信息

  • 1. 北京国网信通埃森哲信息技术有限公司,北京 100052;沈阳建筑大学计算机科学与工程学院,沈阳 110168
  • 2. 北京国网信通埃森哲信息技术有限公司,北京 100052
  • 3. 国网河北省电力有限公司雄安新区供电公司,河北雄安 071000
  • 折叠

摘要

开展电力物资质量检测能够避免不合格的电网材料与设备在电网中投入运行,从而在基础层面确保电网的运行安全性.针对当前的电力物资质量检测方法存在的误差偏高的情况,本研究基于贝叶斯概率更新过程,设计了一种新的电力物资质量检测方法.首先,利用Abaqus CAE有限元仿真软件实施电力物资的有限元建模.然后,定义电力物资的损伤状态等级及其对应的损伤情况,通过向初始有限元模型不断施加载荷模拟电力物资各损伤状态,获得电力物资结构的时域响应,完成电力物资各损伤状态下的数值模拟.最后,基于贝叶斯概率更新过程,设计电力物资结构物理参数变化识别方法,从而完成物资质量检测.实验测试结果表明:该方法能够实现电缆、断路器、矩形母线固定金具的质量检测,识别结果的相对误差整体较低,最低仅为1.62%,同时其变异系数整体较低,最低仅为5.96%,说明该方法能够有效实现设计预期.

Abstract

Conducting quality inspection of power supplies can prevent unqualified power grid materials and equipment from being put into operation in the power grid,thereby ensuring the operational safety of the power grid at the basic level.In response to the high error in current power material quality inspection methods,this study designs a new power material quality inspection method based on Bayesian probability update process.Firstly,use Abaqus CAE finite element simulation software to implement finite element modeling of power supplies.Then,define the level of damage status and corresponding damage situations of power supplies,simulate each damage state of power supplies by continuously applying loads to the initial finite element model,obtain the time domain re-sponse of the power supply structure,and complete the numerical simulation of each damage state of power supplies.Finally,based on the Bayesian probability update process,a method for identifying changes in physical parameters of power material structures is de-signed to complete material quality testing.The experimental test results show that this method can achieve quality detection of ca-bles,circuit breakers,and rectangular busbar fixing fittings.The relative error of the identification results is generally low,with a minimum of 1.62%.At the same time,its coefficient of variation is generally low,with a minimum of 5.96%,indicating that this method can effectively achieve the design expectations.

关键词

有限元建模/数值模拟/贝叶斯概率更新/电力物资质量检测

Key words

finite metamodeling/numerical simulation/bayesian probability update/quality inspection of power supplies

引用本文复制引用

出版年

2024
自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
段落导航相关论文