A Machine Learning Dataset of Artificial Inner Ring Damages on Cylindrical Roller Bearings Measured Under Varying Cross-Influences 核心 · 已核验
atlas:a-machine-learning-dataset-of-artificial-inner-ring-damages-on-c
Dataset in CSV and Python Formats (MATLAB version available here):
This dataset provides a high-resolution, well-annotated collection of vibration measurements from cylindrical roller bearings, both healthy and with artificially induced inner ring damage. It is designed to support machine learning research addressing domain shift by enabling robust evaluation of model generalization across realistic variations in rotational speed, applied load, and mounting position.
Unlike existing bearing datasets, this resource follows a structured experimental design with controlled covariates known to cause domain shifts. It includes 1,151 multi-axis recordings (20 kHz, 60 s) across multiple bearing instances, damage states, and operating conditions.
Optimized for Leave-One-Group-Out Cross-Validation (LOGOCV), the dataset facilitates rigorous assessment of model robustness to unseen conditions. It also includes:
Detailed metadata on testbed setup, damage geometry, and environmental parameters
Transparent labeling of assembly deviations for anomaly detection research
Python scripts for streamlined data loading and segmentation
This dataset is particularly suited for work in robust ML, domain generalization, fault diagnosis, and industrial condition monitoring.
A detailed description of the data can be found at Data Descriptor.
This research was performed in the context of project VProSaar (“Verteilte Produktion für die saarländische Automotivindustrie: Nachhaltig, Vernetzt, Resilient ”) carried out at the Centre for Mechatronics and Automation Technology gGmbH and funded by the Ministry of Economic Affairs, Innovation, Digital and Energy (MWIDE) and the European Fonds for Regional Development (EFRE).
- 落地页
- https://dx.doi.org/10.5281/zenodo.15376390
- 许可证
- CC-BY-4.0 (判读置信:inferred)
- 国内可访问性
-
国内直连:可达 (2026-07-11 检测)
代理通道:可达 (2026-07-11 检测)
检测口径:lychee 双通道单轮探测;「直连超时」表示检测窗口内未完成,系慢或不稳定证据,不构成封锁证据。 - 发布年份
- 2025
- 发布方
- Zenodo
- 设备类型
rolling_bearing- PHM 任务
fault_diagnosis
故障工况
| fault_type: bearing_inner_race_faultinduction: artificially_seeded |
传感器
| sensor_type: accelerometerobserved_property: vibration_accelerationsampling_rate_hz: 20000.0mounting_note: 多轴/多安装位(域偏移实验设计,1151 组 60 s 记录) |
运行工况
| condition_type: rotating_speedis_varying: True |
| condition_type: loadis_varying: True |
溯源(PROV,7 条)
| source_citation: curation/dataset-shortlist-v0.yaml(mech_oam_hub#303)retrieved_on: 2026-07-08asserted_by: automated_harvestnote: 由清单条目初始化的最小候选卡 |
| about_field: description,publisher,publication_year,license_idsource_url: https://api.datacite.org/dois/10.5281/zenodo.15376390retrieved_on: 2026-07-08asserted_by: automated_harvestnote: DataCite REST 元数据回填;仅填空字段,人工值不覆盖 |
| about_field: tasks,equipment_types,fault_conditionssource_citation: facet-batch-01.yamlretrieved_on: 2026-07-08asserted_by: automated_extractionnote: 代理归纳刻面(依据:名称:人工内圈损伤);候选区,晋升需人工核验 |
| about_field: sensors,operating_conditionssource_citation: facet-batch-03.yamlretrieved_on: 2026-07-08asserted_by: automated_extractionnote: 代理归纳刻面(依据:描述:20 kHz/60 s,变速变载变安装位);候选区,晋升需人工核验 |
| about_field: source_citation: 人工核验:zfbin(抽查后委托批准 2026-07-09)retrieved_on: 2026-07-09asserted_by: human_curatorconfidence_level: human_verifiednote: 晋升核心区。首晋升批次 02:KLS-012 满卡(fill=1.00),七批策展逐批用户裁决 + 策展台抽查后委托执行;预检 evidence/KLS-016/02 |
| about_field: china_accessibilitysource_citation: KLS-009 链接健康扫描(lychee 双通道)retrieved_on: 2026-07-11asserted_by: automated_harvestnote: 定期刷新标注,仅覆盖本字段;历史结果以最新扫描为准 |
| about_field: license_idsource_citation: 人工核验:zfbin(三问拍板 2026-07-11)retrieved_on: 2026-07-11asserted_by: human_curatorconfidence_level: human_verifiednote: 人工改写。license 记法归一 cc-by-4.0 → CC-BY-4.0(SPDX 规范 id,ADR-25 清账①,2026-07-11 用户拍板;许可语义不变) |