基于BP神经网络的炉辊传动部故障诊断  

Fault Diagnosis of Furnace Roll Bearing Based on BP Neural Network

作　　者：索瑞悦 牛锐祥 Suo Ruiyue;Niu Ruixiang(Shanxi Taigang Stainless Steel Co.,Ltd.,Taiyuan Shanxi 030003,China)

机构地区：[1]山西太钢不锈钢股份有限公司冷轧硅钢厂,山西太原030003

出　　处：《山西冶金》2023年第9期41-43,共3页Shanxi Metallurgy

摘　　要：为了实现连续退火炉炉辊传动部的故障精准定位,采用BP神经网络诊断炉辊传动部故障。按照采样周期采集炉辊传动部4个测点的振动信号并提取时域指标;构建基于BP神经网络的炉辊传动部的故障诊断模型;将提取的时域指标制作数据集,构建训练模型并验证。实验结果表明:基于BP神经网络的炉辊传动部故障诊断准确率达到了98.11%,有效实现了炉辊传动部故障的精准定位,为炉辊传动部的故障分析提供了思路。In order to accurately locate the fault of the furnace roll transmission department of the continuous annealing furnace,BP neural network is used to diagnose the fault of the furnace roll transmission department.Firstly,the vibration signals of the four measuring points of the furnace roll transmission department are collected according to the sampling period and the time-domain indexes are extracted.Then,the fault diagnosis model of furnace roll transmission department fault diagnosis model based on BP neural network is constructed.Finally,the extracted time domain indicators are made into data sets,and the model is trained and verified.The experimental results show that the accuracy of the fault diagnosis of the furnace roll transmission department based on BP neural network reaches 98.11%,which effectively realizes the fault location of the furnace roll transmission department,and provides ideas for the fault analysis of the furnace roll transmission department.

关 键 词：BP神经网络 炉辊传动部 故障诊断 

分 类 号：TF31[冶金工程—冶金机械及自动化]