MetroPT
收藏资源简介:
MetroPT数据集是由葡萄牙波尔图大学的研究团队创建,旨在为预测性维护提供一个基准数据集。该数据集于2022年收集,包含来自城市地铁公共交通系统的多种传感器数据,如压力、温度、电流消耗、控制信号及GPS信息,总计约10979547条数据。数据集的创建过程涉及从地铁车辆的空气生产单元(APU)中收集连续数据流,并通过GSM网络定期传输至远程服务器。MetroPT数据集主要应用于机器学习算法的评估,特别是在在线异常检测和故障预测领域,以期通过早期故障检测减少运营问题和意外停机时间,从而改变维护模式,从反应性维护转向预测性维护。
The MetroPT dataset was created by a research team from the University of Porto, Portugal, to serve as a benchmark dataset for predictive maintenance. Collected in 2022, this dataset contains various types of sensor data from urban metro public transportation systems, including pressure, temperature, current consumption, control signals and GPS information, totaling approximately 10,979,547 data records. The dataset creation process involves collecting continuous data streams from the air production unit (APU) of metro vehicles, and regularly transmitting these streams to a remote server via the GSM network. The MetroPT dataset is primarily used for evaluating machine learning algorithms, especially in the fields of online anomaly detection and fault prediction, with the goal of reducing operational issues and unplanned downtime through early fault detection, thereby transforming maintenance modes from reactive maintenance to predictive maintenance.




