MagBridge-Battery: A Synthetic Bridge Dataset for Li-ion Magnetometry and State-of-Health Diagnostics

MagBridge-Battery v1.0 is the first dataset to bridge two previously disconnected worlds of battery research: magnetic-field sensing and electrochemical state-of-health diagnostics. Magnetometry is a promising route to non-invasive, contactless battery diagnostics, but progress has been bottlenecked by a structural gap in the available data — the largest open magnetometry archives carry no electrochemical health labels, while the richest state-of-health (SOH) datasets carry no magnetic signatures. MagBridge-Battery closes that gap. Using a bridging procedure that conditions the Mohammadi–Jerschow OSF magnetometry archive on electrochemical labels from the PulseBat dataset, it produces 6,760 labeled magnetic-field signatures for lithium iron phosphate (LFP) cells — enabling machine-learning research on magnetic battery diagnostics that was not previously possible at scale. The release contains 5,600 PulseBat-conditioned grounded samples, 600 synthetic sensor-anomaly samples derived from clean parents (four subtypes: sensor_dropout, calibration_drift, temporal_warp, periodic_interference), and 560 low-voltage Regime-B extrapolation samples. A cell-disjoint, parent-child-leakage-free primary benchmark split is verified to contain zero overlapping cells, zero cross-split parent-child pairs, and zero sample-ID overlap. Four benchmark tasks are defined: SOH regression, second-life classification (cutoff SOH = 0.85), three-class anomaly detection, and four-class anomaly subtype classification. Bridge validity is established through structural sanity invariants, distributional KS tests at grounded anchors, and a controlled label-shuffle ablation that collapses SOH regression from R² ≈ 0.77 to R² ≈ 0 — confirming that the bridge encodes input SOH non-trivially rather than producing label-aligned artifacts. Users are kindly requested to cite both this dataset DOI and the associated paper (see CITING.md in the bundle). Code, paper source, and reference implementations are available on GitHub at https://github.com/SakthiGs/MagBridge-Battery. ARXIV: https://arxiv.org/abs/2605.20240MagBridge-Battery v1.0是一款包含6760条磷酸铁锂（Lithium Iron Phosphate, LFP）电池磁场特征的合成数据集，其生成采用桥接流程：将Mohammadi-Jerschow OSF磁力测量档案基于PulseBat数据集的电化学标签进行条件化处理。本次发布包含5600条经PulseBat数据集标签条件化后的标注基准样本、600条源自干净父样本的合成传感器异常样本（涵盖4个子类型：传感器丢包、校准漂移、时间扭曲、周期性干扰），以及560条低压Regime-B区间外推样本。 该数据集采用无电池重叠、无父子样本泄露的基准划分方式，经验证不存在重叠电池、跨划分父子样本对以及样本ID重复的问题。共定义了4项基准任务：健康状态（State of Health, SOH）回归、二次寿命分类（SOH截断阈值为0.85）、三分类异常检测以及四分类异常子类型分类。 桥接验证环节包含结构合理性不变性检验、基准锚点处的分布柯尔莫哥洛夫-斯米尔诺夫（KS）检验，以及受控标签打乱消融实验：该实验将SOH回归的决定系数R²从约0.77降至约0，证实桥接流程非平凡地编码了输入的SOH标签，而非生成与标签对齐的伪影样本。 恳请使用者同时引用本数据集的DOI与相关论文（详见数据包内的CITING.md文件）。代码、论文源码与参考实现已开源至GitHub平台，地址为https://github.com/SakthiGs/MagBridge-Battery。 ARXIV：https://arxiv.org/abs/2605.20240