高精度多工况H型钢轧机轴承运行故障模拟数据集

该数据集由燕山大学机械工程学院和国家冷轧板带装备及工艺工程技术研究中心的王健、于华鑫、李明、杜金龙和张舒岳团队创建，主要面向H型钢轧机轴承的故障诊断研究，旨在满足智能制造和设备健康监测的需求。数据集基于大型轴类故障实验平台和新型温振复合传感器，通过高采样率多通道数据采集卡采集了正常、内圈故障、外圈故障、滚子故障及复合故障轴承的振动信号。数据集共包含48个CSV文件，数据量为930MB，涵盖了8种健康状态的轴承在不同电机工作频率（2Hz至12Hz）下的振动信号，采样率为51.2kHz。该数据集为轴承故障诊断、智能维护决策及深度学习算法的开发提供了重要数据支撑，推动了轧机轴承故障预测与健康监测技术的发展。

This dataset was created by the team led by Wang Jian, Yu Huaxin, Li Ming, Du Jinlong and Zhang Shuyue from the School of Mechanical Engineering, Yanshan University and the National Engineering Research Center for Cold Rolling Strip Equipment and Technology. It is mainly developed for fault diagnosis research on H-section steel rolling mill bearings, aiming to meet the demands of intelligent manufacturing and equipment health monitoring. Based on a large-scale shaft fault experiment platform and a novel temperature-vibration composite sensor, vibration signals of bearings under normal, inner race fault, outer race fault, roller fault and compound fault conditions were collected using a high-sampling-rate multi-channel data acquisition card. The dataset contains a total of 48 CSV files with a total data volume of 930 MB, covering vibration signals of bearings under 8 health conditions at different motor operating frequencies ranging from 2 Hz to 12 Hz, with a sampling rate of 51.2 kHz. This dataset provides important data support for bearing fault diagnosis, intelligent maintenance decision-making and the development of deep learning algorithms, and promotes the advancement of rolling mill bearing fault prediction and health monitoring technologies.

搜集汇总

背景与挑战

背景概述

该数据集由燕山大学团队创建，用于H型钢轧机轴承故障诊断研究，包含48个CSV文件、共930MB数据，涵盖了正常、内圈、外圈、滚子及复合故障等8种健康状态下的振动信号，采样率为51.2kHz。它为轴承故障诊断和智能维护决策提供了重要数据支持，有助于推动轧机轴承故障预测与健康监测技术的发展。