Thomas Kevin, Rahman Ahasanur, Rohouma Wesam, Ahamed Md Faysal, Shafi Fariya Bintay, Nahiduzzaman Md, Khandakar Amith
Department of Electrical Engineering, College of Engineering, Qatar University, Doha, Qatar.
Department of Electrical Engineering, College of Engineering Technology, University of Doha for Science and Technology, Doha, Qatar.
Sci Data. 2025 Aug 22;12(1):1468. doi: 10.1038/s41597-025-05437-3.
Induction motors are critical to industrial operations but are prone to mechanical and electrical faults. This paper introduces a new dataset for comprehensive fault diagnosis of three-phase induction motors, featuring synchronized multi-sensor data collection. Real-time measurements of vibration, voltage, and current were captured from a 0.2 kW squirrel cage induction motor using high-resolution sensors, with all signals sampled at 50 kHz. Fault scenarios, including phase removal and mechanical misalignments, were simulated to capture diverse motor behaviors. The dataset, organized into ten distinct CSV files covering various operational states, provides a rich resource for developing and testing fault detection algorithms. A Random Forest classifier trained on this dataset achieved an accuracy of 99.82%, demonstrating its suitability for real-time fault diagnosis and predictive maintenance applications. Unlike existing datasets, this collection offers synchronized electrical and mechanical sensor data, enabling advanced cross-sensor fault analysis. The dataset is publicly available and aims to support researchers in advancing machine learning approaches for motor health monitoring.
感应电动机对工业运行至关重要，但容易出现机械和电气故障。本文介绍了一个用于三相感应电动机综合故障诊断的新数据集，其特点是同步多传感器数据采集。使用高分辨率传感器从一台0.2千瓦鼠笼式感应电动机捕获振动、电压和电流的实时测量值，所有信号均以50千赫兹采样。模拟了包括断相和机械不对中在内的故障场景，以捕捉电动机的各种行为。该数据集被组织成十个不同的CSV文件，涵盖各种运行状态，为开发和测试故障检测算法提供了丰富的资源。在此数据集上训练的随机森林分类器的准确率达到99.82%，证明了其适用于实时故障诊断和预测性维护应用。与现有数据集不同，该集合提供同步的电气和机械传感器数据，能够进行先进的跨传感器故障分析。该数据集可公开获取，旨在支持研究人员推进用于电动机健康监测的机器学习方法。
