Sehri Mert, Dumond Patrick
Department of Mechanical Engineering, University of Ottawa, 161 Louis Pasteur, Ottawa, Ontario, Canada.
Data Brief. 2024 Feb 2;53:110144. doi: 10.1016/j.dib.2024.110144. eCollection 2024 Apr.
Induction motors are used in industry as they are self-starting, reliable, and affordable. Applications for these motors include lathes, mills, pumps, power conveyor belts, and commercial electrical and hybrid vehicles. Induction motors have various types of failures, including rotor unbalance, rotor misalignment, stator winding faults, voltage unbalance, bowed rotor, broken rotor bars, and faulty bearings. There is a need for differentiating mechanical faults from electrical fault signals when identifying what part of the motor needs maintenance while using machine learning. Therefore, data collection is essential for electric motor fault diagnosis. The University of Ottawa Electric Motor Dataset - Vibration and Acoustic Faults under Constant and Variable Speed Conditions (UOEMD-VAFCVS) is provided to address this issue. Data from accelerometers, temperature, and acoustic sensors are collected to provide quality electric motor fault data. The dataset includes various induction motor faults useful for time domain analysis. The high-quality data provided by this dataset will help facilitate the differentiation between mechanical faults and electric faults when using fault detection methods, which is a valuable asset for machine condition monitoring.
感应电动机因其能够自启动、可靠且价格低廉而在工业中得到应用。这些电动机的应用包括车床、铣床、泵、动力输送带以及商用电动和混合动力车辆。感应电动机存在多种故障类型，包括转子不平衡、转子不对中、定子绕组故障、电压不平衡、转子弯曲、转子导条断裂以及轴承故障。在利用机器学习确定电动机的哪个部件需要维护时，需要将机械故障与电气故障信号区分开来。因此，数据收集对于电动机故障诊断至关重要。渥太华大学电动机数据集——恒速和变速条件下的振动与声学故障（UOEMD-VAFCVS）就是为解决这一问题而提供的。收集来自加速度计、温度和声传感器的数据，以提供高质量的电动机故障数据。该数据集包括各种对时域分析有用的感应电动机故障。此数据集提供的高质量数据将有助于在使用故障检测方法时促进机械故障和电气故障的区分，这对于机器状态监测是一项宝贵的资产。
